by Sunil Amrith
Nicholson’s successor, James Hornell (1865–1949), devoted years of his life to understanding the fisheries of India’s eastern coast. At the turn of the twentieth century, after a decade working on the isle of Jersey, Hornell traveled to Ceylon to survey the marine fisheries there. From 1908 to 1924, he played a leading role in running the Madras Fisheries Department; he undertook detailed studies of coastal fisheries, on the economy of fishing and the changing composition of the catch; he developed a particular fascination with indigenous fishing vessels along the coasts of the Indian and Pacific Oceans, on which he published over a hundred articles in his lifetime. More than Nicholson, Hornell’s caution about the wholesale technological transformation of India’s fisheries was based on respect for fishers’ traditions and their ways of life. In 1917, Hornell described the daily scene on the shore at Tuticorin, long a center of India’s pearl fishing industry. “There is no wholesale fish market except the beach, there are no companies or large owners controlling each a number of boats, and while there are certainly some fish salesmen and traders, these men seldom or never keep any accounts,” he reported. “The catch is usually thrown in a heap on the beach and the ‘lot’ as it lies is sold by auction—the buyers must appraise its value by eye, and make their bids accordingly.”39 But change was on the horizon; at sea, as on land, new technology and new commercial ambitions strained against the limits that British policies, and British interests, had placed upon them.
IV
The quest for water altered India’s economic balance. For centuries, it had been a matter of common sense that India’s wealth lay in its river valleys—they were the most densely settled, the most intensively cultivated, the most prosperous regions. They had always been the core regions of political power. Wealth and poverty, it so often seemed, were a function of geography. The meteorological divisions of India mapped onto the standard of living. “The most densely populated, and therefore the most fertile regions are those of silt deposit, the population becoming most dense towards the river deltas,” wrote agricultural hydrologist Edward Buck in 1907.40 It appeared an immutable fact of nature—but even by the time Buck was writing, the link between population density and agricultural productivity was no longer so clear. By the 1910s, the largest of Punjab’s Canal Colonies brought more revenue to the government than any other district in India.
Artificial irrigation sparked an agricultural boom in the drier regions of India’s west and south: a boom focused on the production of high value crops for export. Public investment in irrigation poured into regions and crops most likely to bring the state revenue and to bring farmers profit. Punjab and Sind in the northwest, parts of Gujarat and Bombay Presidency in the west, and some parts of Madras—the western region, around Coimbatore, and coastal Andhra in particular—flourished. Those regions still remain India’s most prosperous. The gap in productivity between these favored regions and the rest of the country has grown more marked in the second half of the twentieth century, but the roots of a fundamental reversal in India’s economic geography lie in the water boom of the late nineteenth and early twentieth centuries.
The “ancient” zones benefited far less than the irrigated areas from new markets and new technologies. Riverine Bengal, the fabled wealth of which had drawn in the East India Company in the seventeenth century, now saw a protracted decline in its relative economic position within India. Agricultural productivity in the valleys of Tamil Nadu declined precipitously. A “tide of indebtedness” consumed smallholders in both regions, and also along the Gangetic valley. In both the boom regions and those left behind, the control of water as well as the control of credit concentrated land in fewer hands. Dry regions that had not benefited from irrigation did worst of all. By the 1920s, most agricultural land in India was still not irrigated, it was rain-fed. Already by the early twentieth century, irrigated lands produced four times as much as those that depended on rainfall. A recent survey by economic historians emphasizes the point: for most of India, most of the time, rainfall was the most important factor shaping agricultural output, and yields were probably among the lowest in the world.41
Two large commissions of inquiry in the 1920s—the Royal Commission on Agriculture in India, and the Banking Enquiry Commission, both on the model of the 1901 irrigation commission but on an even larger scale—revealed Indian cultivators’ continuing vulnerability to climatic fluctuations. Irrigation had advanced rapidly, but most Indian farmers did not benefit from it. “Except in the north-west,” the report on agriculture maintained, “the whole country is dependent on the monsoon and all major agricultural operations are fixed and timed by this phenomenon.”42
The main response of Indian cultivators to this vulnerability was to borrow money. The scale of rural indebtedness in India emerged from the inquiry into India’s banking system, a colossal exercise in information-gathering that went province by province, reporting its results in 1930 in dozens of thick volumes. “From the sowing of the seed to the sale of the products,” one of hundreds of witnesses in the United Provinces of North India testified, “it is the indigenous moneylenders and bankers who enable the produce of the villages to be brought to the market.” The monsoon shaped the rhythms of the money market. Cultivators’ need for credit converged on certain crucial times in the season: credit for seed and manure in October and November, after the rains; credit to pay agricultural labor at harvest time. In many parts of rural India there was “an exaggerated alternation of over-work and unemployment.” Account after account from rural India made the same point. “The annual rainfall is scanty and uncertain and irrigation is nominal,” wrote a local official from Meerut district in the United Provinces; and “a peasant proprietor once entrapped by the mahajan [moneylender] can never extricate himself.”43 Many borrowers faced 24 percent compound interest. “Ninety-five per cent of the agricultural classes are in debt,” wrote an administrator in Mathura district, “due to successive failures of crops in this district.”44 Only a few respondents disagreed. The collector of Kaira district in Bombay Presidency told the agricultural inquiry that monsoon failure was not one of the main causes of indebtedness: “The frequency of very bad seasons in which the cultivator would be left completely insolvent is not very great,” he noted.45
The most creative response to the fact that most Indians still depended on the rains for their livelihoods came from J. S. Chakravarti, who worked for years in Mysore. He was a colleague of the great engineer Visvesvaraya; Chakravarti, too, flourished in the southern Indian princely state that was bolder than any part of British-ruled India in its approach to the problems of water. Chakravarti worked for Mysore’s State Insurance Committee for many years, and in the 1910s rose to the position of controller and financial secretary to the government of Mysore. He advocated a system of drought insurance in preference over generalized crop insurance: while the failure of a particular crop could be down to bad practice or neglect on the part of individual farmers, a failure in the rains affected everybody. Where rainfall dipped below a certain proportion, say 35 percent, of average, cultivators would receive a payout. Chakravarti saw rainfall insurance as “intimately connected with three sciences, viz. economics, meteorology, and agriculture.” His starting point was that “Indian agriculture is dependent almost entirely on rainfall” and that the “quantity of rain during the year and its distribution as regards time are almost the only essential factors” in determining the incomes of farmers; absence (or excess) of rain at certain critical periods in the growing season could be devastating. And what Chakravarti called the “rainfall factor” was “uncontrollable by human exertions.” Only the state could carry out an insurance scheme on the scale he envisaged, though he hoped that private providers might eventually enter the market. The case for insurance was clear. “Agricultural insurance will also be famine insurance,” Chakravarti declared, for “under the present circumstances, a famine in India does not generally mean grain-famine, but money-famine, due to enforced unemployment of the
agriculturist.” Chakravarti’s scheme gathered dust; one Indian commentator observed toward the end of the twentieth century that Chakravarti’s scheme was far in advance of what the World Bank had come up with by the early 1990s. The primary reason for this neglect is that a very different approach to mitigating climatic risk emerged in mid-twentieth century India—an approach that emphasized technological solutions to the problem of water.46
Reflecting on the state of India’s development, the members of the Indian Industrial Commission wrote confidently in 1918 that “the terrible calamities which from time to time depopulated wide stretches of country need no longer be feared.” In a monsoon climate, “failure of the rains must always mean privation and hardship,” but it no longer need lead to “wholesale starvation and loss of life.”47 This conveyed a strong sense that something fundamental had changed in India over the first two decades of the twentieth century. The risk posed by climate had been mitigated by both policy (the early-warning system of the Famine Codes) and by technology (railways and irrigation). As long as India remained predominantly agrarian, some level of risk would remain, but the commissioners envisaged a future in which industrialization would provide new employment and greater security, as India’s population moved from the countryside to the cities. In the 1870s, the idea that famine was inevitable in India prevailed among British administrators. By the 1920s, most observers believed that India had conquered famine. But anxieties about water did not go away.
V
India’s engineers fought to assert their sovereignty over the monsoon; climate science demonstrated, in those same years, that the monsoon moved to planetary rhythms—rhythms far beyond human control. The pioneer of Indian monsoon meteorology in the early twentieth century was a brilliant mathematician and a modest man: “kindly, liberal minded, wide of interest, and a very perfect gentleman.”48
Gilbert Thomas Walker was born in Rochdale, Lancashire, in 1868, the fourth child in a family of eight. He grew up in Croydon, just outside London, where his father was the borough’s chief engineer. Gilbert received a scholarship to the elite St. Paul’s School in 1881, and went on to a distinguished undergraduate career in mathematics at Trinity College, Cambridge. His talents were idiosyncratic. At Trinity he left behind him “the legend of his prowess in throwing boomerangs on the Cambridge Backs”—his study of their aerodynamics marked the beginning of his fascination with the physics of the atmosphere; he acquired the nickname “Boomerang Walker.” In 1890, Walker suffered a breakdown in his health; he spent three summers recovering in Switzerland, where he developed a passion for skating. In time, both hobbies would nourish his insights into the world’s weather. Returning to Trinity as a fellow in mathematics, Walker’s work focused on electromagnetism. Formalizing his fascination with boomerangs, he also wrote an essay on the aerodynamics of sports and games. There was little in his background or experience to suggest that within a few years, he would be director of the Indian Meteorological Department.49
John Eliot was approaching retirement. His models for monsoon forecasts had grown more complex during his years in charge of the Indian weather service, but they were erratic in accuracy and had failed to predict the droughts of 1899 and 1900. Convinced, still, that he was on to something, Eliot searched for an able statistician as his successor, someone who could make sense of the profusion of pressure, temperature, and wind readings dispatched from observatories across the Indian and Pacific oceans.50 Walker’s reputation reached Eliot’s notice, perhaps through shared Cambridge networks; the chance to develop a new field of inquiry, in an unfamiliar land, was attractive to the younger man. Before sailing for India, Walker toured meteorological observatories in Europe and the United States: this was his crash course in the science of weather. Visiting field stations in the Midwest, Walker was especially impressed by the sophisticated techniques deployed by the US Weather Bureau.
Walker arrived in India just three years after the last major famine. Eliot’s failures had dented public and official confidence in meteorology. In 1904, Walker took over as chief. He kept a low profile for the next four years: marshaling resources, recruiting staff, bolstering a skeletal meteorological department. In those early years in Simla—the Himalayan summer capital of the Raj, where British officials rushed to escape the pre-monsoon heat—Walker found time for his two beloved hobbies. He was often spotted flinging boomerangs on Annandale, the only stretch of flat ground in Simla. He designed a low canvas screen to keep the ice cold on Simla’s skating rink—it was so effective that, by January, “the ice was too hard to be skated on with pleasure,” and the rink’s owner asked him to remove it. Walker’s mind leapt to make new connections. Many years later he revealed that his experience with the ice rink had led him to understand the “extreme transparency of the air to heat radiated from the ground during the very dry winter periods” over North India.51
From these beginnings came a vital breakthrough in climate science.
WALKER’S APPROACH TO UNDERSTANDING AND PREDICTING THE monsoon was resolutely empirical. He relied on a vast “human computer”—that is, on the labor of his Indian staff led by Hem Raj—to process the numbers. “The relations between weather over earth are so complex,” Walker felt, that “it seems useless to try to derive them from theoretical considerations”—the monsoon was too complex.52 Rather, Walker sought to amass and analyze as much weather data as he could, from all over the world. This had been Blanford’s challenge: to determine how far the monsoon’s causes as well as its effects were confined to “India and its seas.” Blanford’s first forecasts relied on Himalayan snowfall; while he was increasingly aware of remote influences on India’s climate, he assumed a closed system. Eliot moved to incorporate influences from across the Indian Ocean into his model, but he misunderstood the relationships at work. Walker, with more statistical tools (and more staff) at his disposal, broadened his parameters. His team processed a quantity of data that would have been inconceivable a generation earlier. The numbers told a clear and a new story: they suggested that “the monsoon system extended to a pan-oceanic,” even planetary, scale.53
Walker took aim, first, at the desiccationists. He showed that there was little evidence to support their claim that human activity, and particularly deforestation, had modified India’s climate in the nineteenth century. However much the denudation of forests may affect the climate and the soil moisture of a particular locality, the scale of the monsoon system far outstripped such local influences.54 As his thoughts turned to monsoon prediction, Walker looked west of India. The Nile had long been on the minds of India’s hydraulic engineers, as comparison, inspiration, or competition; Walker turned his attention to the annual Nile flood for different reasons. “Inasmuch as the Nile flood is determined by the monsoon rainfall of Abyssinia,” he wrote, “and as the moist winds which provide this rainfall travel in the earlier portion of their movement side by side with those which ultimately reach the north of the Arabian Sea” so there was a “tolerably close correspondence” between the extent of the Nile flood and the strength of the Indian monsoon. This was “seasonal foreshadowing” at work—a term Walker preferred to the more confident “forecasting.”55
Walker’s statistical prowess paved the way for his most startling discovery.56 Mining data from across the world, Walker noticed that “there is a swaying of pressure on a big scale backwards and forwards between the Pacific Ocean and the Indian Ocean, there are swayings, on a much smaller scale, between the Azores and Iceland, and between the areas of high and low pressure in the North Pacific.” The “Southern Oscillation,” as Walker named it, had a “much greater” influence on “world weather” than the other two.57 He called these pivotal areas of high and low pressure “centers of action.” Walker had identified an inverse relationship of atmospheric pressure at sea level across the Pacific Ocean, measured by readings from stations at Darwin and Tahiti. The usual pattern was for high pressure in Tahiti and low pressure in Darwin, driving the winds from east to west.
The pressure contrast across the Pacific drove the storied westerly “trade winds”; but they were prone to periodic reversals that could last for one or two seasons. This was the mystery at the heart of Walker’s findings. Changes in the location and intensity of the “centers of action” shaped the world’s climate—but what prompted these changes?
Walker’s immediate challenge, as head of the Indian weather service, was to determine how these “centers of action” affected India. The broad contours of the picture were clear by the early 1920s. “Abundant Indian rains,” he wrote,
… tend to be associated with low pressure in India, Java, Australia and S. Africa; with high pressure in the central Pacific Ocean (Samoa and Honolulu) and South America (Chile and the Argentine); with previously scanty rainfall in Java, Zanzibar, Seychelles and South Rhodesia; and with low temperature in the Aleutian islands.
He sought correlations in time as well as in space; there were “foreshadowings”—in Zanzibar or in the Aleutian Islands—of dearth or plenty in India; Walker probed the “lags” of a season, or two, in the relationships he discovered.58 But the forces at work proved elusive: “I cannot help believing that we shall gradually find out the physical mechanism by which these [oscillations] are maintained,” Walker said in 1918. A few years later he told the audience at his presidential address to the Royal Meteorological Society that “variations in activity of the general oceanic circulation” would likely be “far reaching and important.”59 It would take another forty years for it to become clear just how “far reaching and important” the oceanic dimension really is.