320,000 Tons of Wheat Sailed Out of India While the Deccan Starved

In 1877 the grain kept moving.

Wheat came down from the northern provinces by rail, went into sacks at the docks, and sailed. Across the worst famine years of the century, grain merchants exported a record 6.4 million hundredweight of wheat, roughly 320,000 tons, to Europe. It was the largest wheat export the country had ever sent, and the government had given strict orders that nothing be done to stop it.

Much of the pull came from Britain itself, where poor harvests in 1876 and 1877 had pushed prices up and created a demand that absorbed most of northern India’s surplus.

Far to the south, in the dry uplands of the Deccan and across the Madras Presidency, people were walking to relief camps and dying before they arrived.

Both of those things were lawful. Both were policy. That is the story of the Victorian famines in one image, and it is why historians who have gone through the paperwork keep landing on the same uncomfortable word: man-made.

Nobody agrees on how many died

The famine of 1876 to 1878 began the ordinary way, with the monsoon failing over southern and western India two years running. The drought was real. What happened next was not weather.

The death toll is contested and you should be suspicious of anyone who gives you a single confident number. The low figure, drawn from official returns, is about 5.5 million. Mike Davis argues that this is certainly too low, because it left out the princely states entirely and counted none of the later deaths from high prices and the malaria that followed. The demographer Arup Maharatna puts the toll at around 8.2 million, and some historians go higher still.

That gap of nearly three million people is not a rounding error. It is the measure of how little the administration wanted to know.

Counting was itself a fight. The historian David Hall-Matthews has shown that the arguments inside the colonial government over how many had died, and over how much food a starving person actually needs, were arguments about whether the policy could be defended at all. Get the numbers low enough and the policy looks like prudence.

The rule: relief must not spoil the market

The machinery that turned a drought into a catastrophe was not cruelty for its own sake. It was doctrine.

The ruling economic idea of the day was that governments should keep their hands off the grain trade. If prices rose, merchants would be drawn in, supply would follow the money, and the market would fix the shortage faster than any official could. Buying grain, capping prices, or handing out free food would “interfere,” discourage private traders, and make the problem worse.

There was a second belief bolted onto the first, and it did more damage. Relief that was too comfortable would make Indians dependent on it. Officials of the period called this demoralisation. The fix was to make relief deliberately unpleasant, so that only the truly desperate would take it.

Put those two ideas together and you get a system with a built-in floor under how much help anyone was allowed to receive, no matter how many were dying.

One pound of grain for a day of digging

Sir Richard Temple had been criticised for spending too freely on relief during the Bihar famine of 1873 to 1874. Sent south as famine delegate in 1877, he made sure the charge would not stick twice.

In January 1877 he cut the relief wage in the Madras and Bombay camps. A man doing a full day of hard labour would receive one pound of grain, about 450 grams, plus one anna, a coin worth roughly fourpence. A woman or a working child got slightly less. The work was digging and hauling on roads, canals and railway earthworks, a long day with no shade and no rest.

The ration was later raised in Madras to roughly 570 grams of grain with a little dal, meaning lentils, after enough people had died to make the point.

To see the scale of it, hold it against the ration books of the twentieth century. Mike Davis, whose Late Victorian Holocausts is the best-known account of these years, sets the Temple ration side by side with the diet recorded at Buchenwald, and the Temple ration comes out lower. It is a jarring comparison, it belongs to Davis rather than to any neutral ledger, and it has been argued over. It also fixes the size of the thing in a way that grams on a page do not.

And you had to walk to it. Relief works were placed where the government wanted infrastructure built, not where the hungry already were. Families crossed districts on foot to reach a camp that would then pay them less food than the walk had cost them.

Viceroy Lord Lytton backed Temple throughout. In the same month that the ration was cut, Lytton was in Delhi proclaiming Queen Victoria Empress of India.

The doctor who did the arithmetic

The angriest versions of this story leave out the part that follows, and leaving it out weakens the case rather than strengthening it. People inside the system said this was wrong, in writing, at the time.

W. R. Cornish, the Sanitary Commissioner for Madras and a working physician, went through the numbers and concluded that Temple’s ration could not keep a labouring adult alive. He argued for a minimum of about 680 grams of grain, plus vegetables and some protein, particularly for anyone doing heavy work. He was overruled.

The journalist William Digby, who ran the private relief campaign in Madras, put it more bluntly when the government declared the famine under control. “A famine can scarcely be said to be adequately controlled,” he wrote, “which leaves one fourth of the people dead.”

They lost. That is the finding, and it matters. This was not a system with no other options on the table. The other options were tabled, costed, and rejected because they cost more.

The insurance that was not there

After the dying stopped, the government did what governments do. It appointed a commission.

Lytton’s own government had already set up the Famine Insurance Grant in 1878: roughly £1.5 million a year, to be put aside out of revenue so the next famine could be met without borrowing. The Indian Famine Commission, appointed that same year and reporting in 1880, blessed the arrangement and led on to the Provisional Famine Code of 1883, a first template for the provincial codes. Davis judges the report itself toothless, and the codes that followed inconsistent.

It was never a vault with money in it. It was an accounting promise to apply surplus revenue, and surpluses have a way of being needed elsewhere. By 1887 a member of the Commons was asking whether the fund had effectively ceased to exist, having gone to pay for the war in Burma. The government denied the Burma charge and then conceded the substance of it: there was no surplus for 1887 to 1888 out of which the famine grant could be paid.

The famines came back regardless. In 1896 to 1897 and again in 1899 to 1900, drought returned and millions more died, this time with a Famine Code in force. Relief broke down worst in the Central Provinces, where 500,000 died in 1900 alone, and worst of all for forest and tribal communities.

What the charge is, and what it is not

The honest version of this argument is narrower than the internet version, and it survives contact with the evidence.

The droughts were real. The monsoon did fail, repeatedly, driven by the same global climate swings that produced famine in China and Brazil in the same years. No British official invented the weather.

Nor was any of this a plan to kill Indians. There is no order to that effect because it was never needed. What the record shows instead is a long series of decisions in which someone weighed Indian lives against the budget, against economic doctrine, and against the smooth running of the export trade, and Indian lives came second every time.

That is the charge, and the export figure is its most checkable piece. A country in which millions were starving went on shipping grain out, and in the worst year it sent more wheat abroad than it ever had. That wheat came largely from northern provinces whose harvests in 1874 to 1876 had been good. When the drought reached those provinces in 1877, the reserves that should have carried them had already been sold abroad, and at least 1.25 million more people died there. The food had existed, the railways to move it existed, and it went the other way.

Sixty-five years later, under a different viceroy and in the middle of a world war, the same logic produced the Bengal famine of 1943, where roughly three million died in a province that had grown enough to feed itself.

A famine is usually described as something that happens to people. In British India it was closer to something that was decided about them, in rooms where the arithmetic was done carefully and the wrong things were being counted.