Taming the Anarchy

Groundwater Governance in South Asia

Tushaar Shah

• Publisher: Routledge
• Year: 2010
• Topic: Politics & International Relations
• eBook ISBN:
  9781136524028
• Subtopic: Development Economics
• Index: Politics & International Relations

In 1947, British India-the part of South Asia that is today's India, Pakistan, and Bangladesh-emerged from the colonial era with the world's largest centrally managed canal irrigation infrastructure. However, as vividly illustrated by Tushaar Shah, the orderly irrigation economy that saved millions of rural poor from droughts and famines is now a vast atomistic system of widely dispersed tube-wells that are drawing groundwater without permits or hindrances. Taming the Anarchy is about the development of this chaos and the prospects to bring it under control. It is about both the massive benefit that the irrigation economy has created and the ill-fare it threatens through depleted aquifers and pollution. Tushaar Shah brings exceptional insight into a socio-ecological phenomenon that has befuddled scientists and policymakers alike. In systematic fashion, he investigates the forces behind the transformation of South Asian irrigation and considers its social, economic, and ecological impacts. He considers what is unique to South Asia and what is in common with other developing regions. He argues that, without effective governance, the resulting groundwater stress threatens the sustenance of the agrarian system and therefore the well being of the nearly one and a half billion people who live in South Asia. Yet, finding solutions is a formidable challenge. The way forward in the short run, Shah suggests, lies in indirect, adaptive strategies that change the conduct of water users. From antiquity until the 1960's, agricultural water management in South Asia was predominantly the affair of village communities and/or the state. Today, the region depends on irrigation from some 25 million individually owned groundwater wells. Tushaar Shah provides a fascinating economic, political, and cultural history of the development and use of technology that is also a history of a society in transition. His book provides powerful ideas and lessons for researchers, historians, and policy

CHAPTER 1

THE HYDRAULIC PAST:
IRRIGATION AND
STATE FORMATION

Irrigation makes you free; irrigation makes you rich.

— Document dated 1243, Catalonia, Spain

Over the millennium past, the history of South Asian irrigation has undergone three distinct yet overlapping phases: the era of adaptive irrigation, up to 1800; the era of constructive imperialism, from 1800 to 1970; and the era of atomistic irrigation, from 1970 to date. This opening chapter attempts a broad-brush narrative, based on a review of the literature, of the evolution of South Asian irrigation over these three eras, as the historical backdrop against which the current problem needs to be viewed.

Accounts of the evolution of irrigation in India are often wholly preoccupied with the great works of kings, overlords and colonial engineers. From the Sudarshan reservoir, built and rebuilt in Saurashtra by Mauryans, to the 2,000-year-old overflow irrigation systems built by the Chola kings in Bengal and recreated in Tanjore, to the irrigation systems of Zainul Abidin in Kashmir, the ancient earthen dams on Cauvery and Vaighai in the south, and later the great canal systems extended from Jamuna by the Delhi sultans and renovated from time to time by the Mughals, the Porumamila tank of Cuddappah in the Vijayanagar empire and the ancient tank cascades of Sri Lanka—in recounting these accomplishments, history textbooks seldom tire of the audacious irrigation adventures of ancient and medieval India's kings and overlords.

Indeed, preoccupation with these large, state-constructed irrigation works led great thinkers like Marx to regard centrally managed irrigation as the defining feature of the “Asiatic mode of production” and ancient and medieval Indian society and politics. “One of the material bases of the power of the state over the small disconnected producing organisms in India,” said Marx famously, “was the regulation of the water supply. The Mohammedan rulers of India understood this better than their English successors” (Marx cited in Habib 1999, 297, n. 47).

Unlike Deakin, who had firsthand experience of the Indian irrigation scene, Marx's inferences were from secondhand accounts. However, the idea that irrigation dictated politics and early state formation in India, China, and other eastern civilizations continued to hold sway decades after Marx. The imperatives of managing large-scale irrigation works became the leitmotif of Karl Wittfogel's once hugely influential book, Oriental Despotism: A Comparative Study of Total Power (1957). In irrigation, Wittfogel was searching for the answer to an epochal question: why, through the first 1,500 years after Christ, eastern absolutism was so much more comprehensive and oppressive than in western societies. In understanding the harshest forms of total power that oriental despotism represented, Wittfogel, like Max Weber and Karl Marx, was particularly impressed by two features of oriental societies: the large waterworks maintained for irrigation and communication, and the fact that the government was often the biggest landowner.

According to Wittfogel, construction and maintenance of large irrigation systems—required in the arid Orient but not in temperate Europe— necessitated a political and social structure capable of forceful extraction of labor, which led to despotism¹ and a different social and political organization in China, India, Egypt, Yemen, Ceylon, and Mesopotamia. Hydraulic management necessitated a structured bureaucratic organization under centralized control, and the building and maintenance of irrigation systems forced people to become unequal leaders. As the systems grew, leadership was required to build new canals, maintain existing ones, and ensure efficient distribution of water.

How accurate and valid were Wittfogel's and Marx's characterization of irrigation as the central determinant of state formation and “oriental despotism?” In Indian history, evidence of an irrigation bureaucracy with responsibility for measuring land and regulating water supply at sluices is found only in Mauryan kingdoms and, to some extent, in the reign of Firoz Shah Tughlak, who built the first Jumna canal (Randhawa 1982). India had some great ancient irrigation works; however, much farming in ancient and medieval India was, in Wittfogel's terms, hydro-agriculture with a strong role of village communities rather than hydraulic farming under state domination. The great works were constructed at disparate times over the course of 2,000 years; it is unlikely that at any given point in time, these were a significant presence in a typical Indian peasant household. Farming here was organized as a family enterprise, whereas water mobilization and distribution had a strong tradition of local community management. The compilation by the Delhi-based Centre for Science and Environment in Dying Wisdom (Agarwal and Narain 1997) of myriad small, indigenous water harvesting and irrigation systems in different parts of South Asia supports the view that decentralized, community-based hydro-agriculture, not centralized hydraulic farming, dominated the region. Around 1800, water was being artificially applied to some six million ha in India; of that land, perhaps no more than a few hundred thousand ha depended on works centrally controlled by the state through an elaborate bureaucracy. Irfan Habib, the eminent historian of Mughal India, suggests, “Despite the urgings of Marx, sufficient evidence does not exist to warrant the belief that the state's construction and control of irrigation works was a prominent factor in the agrarian life of Mughal India” (Habib 1999, 297).² However, much evidence indicates that a tradition of community-level cooperation in mobilizing and managing water for flow irrigation—superimposed over the family as a farming unit—was a prominent aspect of agrarian life in Mughal and earlier periods.³

Discredited as Wittfogel's thesis remains today,⁴ his analysis offers deep and powerful insight into the gradual but persistent atrophy of present-day public and community gravity-flow irrigation in South Asia, and perhaps elsewhere in the developing world, as we examine in Chapter 3.

Irrigation in Mughal India and Earlier

Although South Asia, along with China and Europe, were the most densely populated regions of the world in 1700 (Klein and Ramankutty 2004), farming land went abegging here until 150 years ago. Villages tended to settle at sites hydraulically most opportune for stable and productive agriculture. Riverine areas that were not so low as to be flooded were preferred. The popular notion that civilizations formed around watercourses aptly described settlement patterns of those times.

The state as well as the people lived off the land. Since farming was the primary source of government revenue, the intercourse between the state and the peasants was broad as well as deep, and the apparatus for land revenue administration substantial and prominent. Peasants had few property rights, if any. In Mughal India, no one could sell, mortgage, or bequeath land.⁵ All land belonged to the emperor, who was also the heir to every subject. Commonly, land was held by the village community, and land went back to the community if its tiller died heirless. Rights were given only to till the land and to collect the emperor's share of the produce of land, which except during Akbar's rule was generally higher than one-third, closer to one-half, of the output.

Helping the peasant produce as large a surplus over his subsistence needs was a prime motive of the state right up to the colonial period.⁶ The desire to maximize land revenue is what created incentives to invest in irrigation, not any interest in improving the welfare of the peasant, as today. The South Asian village was far less “settled” and permanent than we know it today; and begar, the institution of forced labor, was widely prevalent. When demand for land tax and begar from the ruling classes became excessive, or during times of war, peasants would desert their lands en masse, depopulating whole villages and migrating with their cattle to forests or hills or to neighboring provinces where revenue demands were lower.⁷ The mobility of peasants was the only check against unbridled rent-seeking by overlords, who had no use for territory without peasants. Hard pressed to keep peasants on their lands, feudal lords sometimes used force but were often more successful by offering wet lands in river valleys or along the banks of the rivers, or by helping (or coercing) peasant communities to build embankments, or bunds, and channels that offered a modicum of water control.

This suited the adaptive strategies of farming communities, and agriculture as well as human settlements intensified and grew in regions that offered natural water control, typically in well-drained wetlands. As Ludden (1999, 49) points out,

In Punjab, as in general through out the northern basins, the long term geographic spread of intensive agriculture moved outward from places where drainage is simpler to harness to places where more strenuous [water] controls are necessary … from naturally wetter into drier areas; where as in flood plains and humid tropics, it moved initially from higher and drier parts of the low lands into the more waterlogged areas at the river's edge. Everywhere, agriculture also moved up river valleys into the highlands. In the Yamuna-Ganga basin, the general trend of expansion of intensive agriculture has been from east to west and upland from low lands; and in the Punjab, from north-east to south-west.

There must have been little by way of pure dryland or rain-fed farming in the sense we understand today—that is, crop cultivation based on rainfall and soil moisture alone in arid and semiarid environments where extensive pastoralism was the customary livelihood. But there was also little of the kind of water control that modern irrigation implies, with its well-managed canals and tube wells. Like basin irrigation that dominated Egyptian agriculture, the inundation or overflow irrigation in Mughal India is best understood as minimalist, adaptive, nonintrusive irrigation that entailed minimal human footprint on the basin hydrology. This involved spreading river floods over nearby arable lands, allowing the water to soak in, and then draining them. As Agnihotri (1996, 42) suggests for northwestern India, farming involved “maximizing the utilization of sailab (floods) as well as digging up watercourses and making minor bunds. The idea being … to raise the area as well as the productivity of riverine cultivation.”⁸ The ahar-pyne systems in eastern India and water tanks of southern India and Bengal had the same purpose. Upon his first visit to northern India in the early sixteenth century, Babur, the founder of the Mughal Empire, was surprised that people left rivers alone but used shallow wells and tanks for irrigation (Bagchi 1995). This was, however, less the case farther east, where floodwaters were used extensively to grow crops.

Wells played an important though supplemental role in water-managed agriculture in northern and western India. There is extensive evidence of the use of open-lined and unlined wells in Vedic literature primarily for domestic water needs but also for supplementary irrigation. The Satwahanas introduced ring wells—dug wells whose shafts were lined with concentric terracotta or sometimes wooden rings—for domestic use and brick wells for irrigation use. Well irrigation, and its productivity-enhancing potential, was firmly established in the revenue calculus of ancient and medieval rulers. Well construction was actively encouraged through incentives and tax remissions, but lands irrigated with wells were also assessed at a higher rate than rain-fed lands from the time of Arthashastra (third century B.C.) to the Mughal rulers during the sixteenth to eighteenth centuries A.D. and, still later, during the colonial era (see, e.g., Hardiman 1998). Wells were preferred because earthen dams were easy prey to marauding armies (Ludden 1999). The significance of well irrigation was growing, in quantitative terms, during the Mughal as well as the colonial eras. Yet as Agarwal and Narain (1997, 285) note, “… little else is known about the long term history of wells because they were family enterprises, and their fortunes went unrecorded.” We may surmise that around two million ha of farmland was irrigated by wells around 1800, and the total quantity of ground (and soil) water lifted during a typical year was around eight to 10 km³. The number of bullock pairs and the human effort needed to irrigate larger areas than this using leather buckets would be too large for the available peasant labor.

Figure 1.1 Traditional water-lifting devices used in India

For centuries, the use of well irrigation was limited by the laborious technologies for lifting water: the leather bucket (charasa), the swing basket, and the bamboo basket (dhenkuli) (see Figure 1.1). The introduction of the Persian wheel (sakiya, or rahat) by West Asians to northern India's peasantry hastened the spread of well irrigation (Hardiman 1998).⁹ A system powered by oxen driven in a circle, the Persian wheel became the universal mode of lifting water from wells or canals in many northern and western parts of South Asia (Islam 1997). Abu-l-Fazl asserted that “most of the province of Lahore is cultivated with well-irrigation,” an assertion repeated by a chronicler of the late seventeenth century who himself lived in that province (Habib 1999, 29).

The Persian wheel was widely preferred because it could be operated with four or six bullocks whereas the charasa required six or eight. Moreover, even a child or a woman could operate the Persian wheel; the charasa required two able-bodied men, one to drive the bullocks and the other to empty the bucket. The Persian wheel could also be adapted to lift different volumes of water and work effectively at different depths of water level. As a result, the Persian wheel commanded a larger area and was more widely used than other devices (Islam 1997). Four major improvements occurred in the Persian wheel during the nineteenth century: the wooden frame was replaced by iron; the heavy earthen buckets were exchanged for lighter aluminum ones; the use of metal roller bearings reduced friction and increased automatic lubrication of moving parts; and finally, toward the end of the nineteenth century, animal power was replaced by diesel engines and electric motors (Islam 1997).

Most irrigation wells in northern India were kuchcha, or earthen, and made afresh every year, which suggests that groundwater levels remained at predevelopment levels during most parts of a year with a normal monsoon season. In the early nineteenth century, as statistics of irrigated areas began to be compiled, it was striking that well digging increased during drought years, as it does even today. Wells were used more as insurance against drought or a dry spell. In years of successive drought, however, the falling water levels severely tested the power of manual lifting devices.

Collective action at the community level, common in flow irrigation, was widespread in well irrigation, too. The groundwater institutions we find in South Asia today were incipient in the early years of the nineteenth century and fully operational in its later decades. Jointly owned wells in nineteenth century Punjab (Islam 1997) operated much as tube well companies of North Gujarat (Shah and Bhattacharya 1993) and Punjab (Tiwari 2007) do today, as well as the wara bandi (rotational) system in canal irrigation systems in the Indus basin. Dividing waters among owners was done in proportion to shares owned. Each share entitled the bearer to a certain number of waris (turns), each wari divided into a prahara (one-eighth of a day) of three hours. The number of shares were usually in proportion to the holder's land in the command (Islam 1997). Far from the complex, state-controlled irrigation machine that Marx attributed to the Mughal Empire, water-managed agriculture involved local cooperation and community rulemaking and enforcement. Historians are unanimous in noting the spread and significance of well irrigation in northwestern and western India as a tradition extending from West Asia.¹⁰ Even with major public investments in canal irrigation during the second half of the nineteenth century, private capital investment in well irrigation remained substantial in these parts.¹¹

However, we seldom hear about irrigation by wells on a large scale in eastern, central and southern India during the Mughal as well as the colonial era, except perhaps in the Coimbatore and Tanjore areas in Madras, where wells acquired some importance in the late nineteenth century (Roy 2000). By far the best account of irrigation in peninsular and eastern India around 1800 comes from Francis Buchanan, a medical officer on the rolls of the East India Company. Irrigation by wells is conspicuous by its absence in Buchanan's travelogue, which is otherwise detailed and incisive in describing agrarian conditions, including the means for irrigation. In today's Tamilnadu, Karnataka, Kerala, and parts of Andhra Pradesh, which he toured extensively in the early nineteenth century, Buchanan reported on vibrant “water-managed”¹² agriculture around tanks and reservoirs but found well irrigation as rare as tanks were ubiquitous (Dutt 1904).¹³ In the hard-rock geology of peninsular India, digging wells was f...

Table of contents

• Front Cover
• Title Page
• Copyright
• About Resources for the Future and RFF Press
• Resources for the Future
• Contents
• About the Author
• Preface
• Introduction
• 1. The Hydraulic Past: Irrigation and State Formation
• 2. Rise of the Colossus
• 3. The Future of Flow Irrigation
• 4. Wells and Welfare
• 5. Diminishing Returns?
• 6. Aquifers and Institutions
• 7. Can the Anarchy be Tamed?
• 8. Thriving in Anarchy
• Endnotes
• Glossary of Hindi and Other Terms
• References
• Index