Moreover, thousands of chowkidars were unpaid in late 1941, and given the progressive breakdown in civil administration as war reached India’s borders and the Quit India movement erupted, there is no reason to suppose that they were paid in 1943. The chowkidar’s salary of 6 rupees a month—plus a “special bonus” of an eighth of a rupee for helping with cyclone relief—was in any case too low for his family’s survival at the prices that prevailed during the famine. Recall that Giribala Malakar of Kalikakundu said that her first husband, a chowkidar, had carried relief materials during the famine and had still starved to death. It is
inconceivable that all through 1943 these functionaries were faithfully discharging their manifold duties to the empire. When the famine inquiry committee asked relief commissioner Olaf Martin about the registration figures, he replied, “I know that they are exceedingly bad.”
8
It is even possible that the figures supposedly compiled by the chowkidars were actually filled in by higher-level officials who had little idea of the toll the famine took and perhaps little interest in recording a disturbingly high number of casualties. After the famine Richard Symonds, a charity worker, was employed by the governor of Bengal to help with relief and rehabilitation. In an effort to learn the number of children orphaned by the famine, he sent forms around to the district offices. On a subsequent tour he discovered why the results he’d gotten had made little sense: “At the very bottom of the Government pyramid, the circular would come to an officer who might be less concerned with accuracy than with the consequences for his career.” One functionary might suppose that a small number of orphans would speak to his zealousness in supplying relief; another might figure that a large number of orphans would lead to generous funds arriving for their support. In sum, the figure either man would put down would be fictitious.
9
Demographers Tim Dyson and Arup Maharatna noted a peculiar pattern in the registration data for West Bengal (the piece of Bengal province that went to India after the partition). During the years 1941 to 1946, the proportion of deaths in certain districts remained exactly the same—a sign that the numbers had been manufactured from the figures for undivided Bengal. Had village-level registration data been available, such manipulation would not have been necessary.
10
Be that as it may, the famine commission declared that “there was no universal breakdown in 1943 in the system of recording deaths.” After introducing a correction for the usual shortfall in death registrations and for roadside deaths, it concluded: “The number of deaths in excess of the average in 1943 was of the order of one million—that is, some 40 percent, in excess of the officially recorded mortality. We have found no valid reason for accepting estimates in excess of this figure. On the other hand, the high excess mortality in 1944 must be added
to the toll of mortality. On this basis we must conclude that about 1.5 million deaths occurred as a direct result of the famine and the epidemics which followed in its train.”
11
Diverse authors have applied equally diverse corrections to the raw numbers supposedly collected by the chowkidars to obtain other estimates for the mortality. Economist Amartya Sen took the registered deaths for West Bengal, extended them to East Pakistan (nowadays Bangladesh), and applied corrections to get around 3 million for the famine toll. Dyson and Maharatna used figures that they unearthed for undivided Bengal, argued that the system of data collection had remained essentially intact during the famine, and got 2.1 million.
12
Intriguingly, historian Paul Greenough discovered that a careful statistical survey of famine mortality had been presented to the famine commission but was never published or publicized. Calcutta was home to one of the world’s foremost statisticians, Prasanta C. Mahalanobis, who had founded the Indian Statistical Institute in 1931. By means of exploratory projects in Bengal, Mahalanobis had developed the sample survey, which permits reliable results for the many to be inferred from those for the appropriately chosen few. To that end he had devised pilot surveys (preliminary studies that help refine the design of the final, exhaustive one), overlapping samples, and other means of reducing error. Mahalanobis would go on to become a Fellow of the Royal Society and to run the United Nations statistics bureau.
13
In late 1943 and early 1944, Mahalanobis and his team designed and carried out an elaborate survey, at government expense, of the famine victims. The researchers interviewed the members of 13,358 households in a representative spectrum of districts and villages to estimate a total number of deaths during 1943 of 3.1 million. (This is not the famine toll, as will presently become clear.) The death rate in Bengal in 1943 came out to be 5.3 percent. The worst-affected subdivision was Bhola in the southeast, with a death rate of 14.79 percent.
14
The survey had its own shortcomings, but these were precisely defined, allowing for a more solid grasp of the problem than the squishy registration data would allow. To begin with, Mahalanobis did not
include the deaths of children less than one year old because the respondents were unreliable on that figure. Second, the survey depended on the reports of relatives, and so could not account for individuals who were missing—some of whom must have died—or for families that had perished in their entirety. (In Kalikakundu at least one family, consisting of Behari and Duari Das and their teenage son, died out.) Third, Mahalanobis did not get to repeat the survey to account for deaths from famine-related disease during 1944 and later. Fourth—and this is the most difficult problem—in order to calculate the famine toll, one has to subtract from the total number of deaths those that would have occurred in the absence of famine.
The “normal” rate of death is very hard to determine, because the death rate varies from time to time and from place to place. The mortality rate for India as a whole is believed to have been 2.1 in 1942, but no data specific to Bengal are available. Compared to the politically favored northwest, the state of nutrition and health in Bengal was miserable, and Mahalanobis assumed a normal death rate of 4 percent for the province, based on the census of 1931.
15
One can gauge the applicability of this death rate to the early 1940s by reviewing how mortality in Bengal might have evolved during the 1930s. The rate of land sales offers strong evidence that economic distress increased all through that decade. (This kind of data is reliable because land sales, unlike deaths, had to be registered in order to take effect.) The number of land sales in Bengal in 1929 had been 79,929, but the frequency of such transactions increased steadily so that in 1938 the figure was three times as high. The next year—when the war began—the sales doubled, and they continued to rise until in 1942 the number was 749,495. In 1943, the famine year, the figure doubled again, to 1,532,241. Since an owner parted with land only as a last resort—to save his life and that of his family—these figures indicate that the suffering in Bengal, and in all likelihood the mortality rate, increased throughout the depression and war years and reached a peak with the famine.
16
The easiest way to begin estimating the number of famine deaths is to picture a pyramid representing a pile of corpses that towers over a
flat expanse—as would be the norm. But with the 1943 Bengal famine, one is forced to try and separate a peak from the mountain upon which it sits. Where should a determination be made between the two? What level of mortality should be regarded as normal? It becomes as much a matter of inclination as of science.
Beginning with Mahalanobis’s figure for the total number of deaths in 1943, Greenough corrected for infant deaths (which were 18 percent of the deaths in normal times) and subtracted the estimate of normal deaths provided by Sen to estimate a famine toll of 2 million for 1943. To that he added Sen’s estimate of mortality during 1944 and 1945 to get between 3.5 and 3.8 million as the number of deaths attributable to famine.
17
A more consistent approach is to stick with mortality rates for the entire calculation. Mahalanobis’s mortality rate of 5.3 percent, when corrected for infant deaths, yields a total mortality rate of 6.5 percent for 1943. The population of Bengal in January 1943 was 61.8 million, which gives a total of 4 million deaths for 1943. From this figure must be subtracted 2.5 million baseline deaths (calculated by using the normal death rate of 4 percent that Mahalanobis preferred). That gives 1.5 million for the famine toll in 1943. Doubling this figure, because death registrations were roughly symmetric around December 1943, provides a famine toll of around 3 million.
18
None of these estimates for famine mortality account for starvation deaths during 1942—although in coastal Bengal, scorched earth and cyclone had caused the famine to begin by at least the last quarter of that year. And it remains unclear how, if at all, Mahalanobis accounted for those individuals and families who had simply vanished. On the other hand, it is possible that the normal death rate was even higher than 4 percent, which would tend to reduce the famine toll. One thing is clear: the figure of 3 million does not include all fatalities from shortage of food, because deaths from malnutrition were undoubtedly occurring even in so-called normal years. If for comparison we were to use the death rate of 2.1 percent that was the norm for India (rather than Bengal) in 1942, the famine toll would be 5.4 million. Nor do
these figures include famine deaths in the rest of the rice belt, such as in Orissa and Madras.
IN 1949, A session of the Geneva Convention extended the guidelines for civilized warfare and included a prohibition against starving civilians in occupied territories. “To the fullest extent of the means available to it, the Occupying Power has the duty of ensuring the food and medical supplies of the population; it should, in particular, bring in the necessary food-stuffs, medical stores and other articles if the resources of the occupied territory are inadequate,” the convention declared. In 1977, additional protocols strengthened the injunction against starvation, prohibiting the removal or destruction of “objects indispensable to the survival of the population.” If the application of an economic blockade resulted in starvation, the convention further mandated the provisioning of essential supplies to the civilian population. In fact, depriving civilians of an occupied territory of vital foods and failing to supply them with adequate relief constitute war crimes as understood today.
19
Given that the War Cabinet enjoyed absolute control over shipping to and from India, that tens of thousands of nationalists were in prison, that in the estimation of military intelligence India was “an occupied and hostile country,” and that the War Cabinet spoke for Indians—by, for instance, turning down offers of grain on their behalf—India was presumed by the United Kingdom to be an occupied territory. If such provisions protecting civilians had been in place before the war, the Denial Policy and the failure of His Majesty’s Government to relieve the famine could conceivably have been prosecuted as war crimes.
ALTHOUGH THE ACTIONS of the War Cabinet can be traced with some accuracy (mainly through documents of the Ministry of War Transport), its motives for denying adequate grain to India in the summer and fall of 1943 remain too various and intertwined to tease apart. One reason it chose not to relieve the famine derived from its determination to meet the target of 27 million tons of civilian imports for the United Kingdom. To Cherwell, at least, that meant no ships could be
released from the import program. A second reason was the Balkan stockpile, close to Churchill’s heart, and also close to Cherwell’s because that reserve could take some of the pressure of feeding liberated Europe off the U.K. stockpile. A third reason appears to have been the avoidance of embarrassment, as in having to admit to American officials that the British Empire controlled enough ships and grain to send substantial relief to a colony imperiled by hunger.
Saving face might seem to be a peculiarly trivial reason for permitting a famine to run its course, but perhaps it was not too trivial, given that the English government deemed the lives of Bengalis to be inconsequential. Churchill’s broad-brush loathing of the natives might have added impetus to the other rationales for failing to aid them, as might have the continued defiance of Subhas Chandra Bose, who was wildly popular among Bengalis. (The mere existence of the Indian National Army was a source of humiliation to the British, because it advertised the fact that armies of subjects in British colonies had chosen to fight alongside the Japanese, whereas the reverse was true in the Philippines, an American possession.) The War Cabinet’s shipping assignments made in August 1943, shortly after Amery had pleaded for famine relief, show Australian wheat flour traveling to Ceylon, the Middle East, and southern Africa—everywhere in the Indian Ocean area but to India. Those assignments suggest a will to punish.
Fiercely evident through almost every War Cabinet discussion on India in 1943 is Churchill’s inchoate rage: rage because the colony was slipping out of his grasp, and rage because in violation of the laws of nature and man it had turned from being a debtor to a major British creditor. As late as 1947, Churchill would write in a draft of his war history that India was “the greatest war profiteer.” (A similar charge could be made of His Majesty’s Government. It stockpiled food during famine, and by selling such liquid assets “must have made a considerable profit on the rising market after the war,” according to historian R. J. Hammond.) Churchill and Cherwell were convinced that for the United Kingdom to owe money to India was a “monstrous” injustice, and that the colony should be forced to bear the entire burden of war expenses.
All the evidence points to the prime minister and his closest adviser having believed that Indians were ordained to reside at the bottom of the social pyramid, such that their financial ascendancy as creditors during the war became a source of frustration and fury. Long after India had obtained independence, the Prof would describe “the abdication of the white man” as the worst calamity of the twentieth century—more deplorable than two world wars and the Holocaust.
20