In a study at Stanford University in the spring of 1968, a thirty-five-year-old butterfly biologist named Paul Ehrlich finished typing a short book that the executive director of the Sierra Club, David Brower, had asked him to write in a hurry, in time to influence the upcoming presidential election. Brower had read a magazine article Ehrlich had published the previous year and decided he wanted it as a paperback.

The first sentence of the book was: "The battle to feed all of humanity is over."

The second sentence, slightly less remembered, was: "In the 1970s and 1980s hundreds of millions of people will starve to death in spite of any crash programs embarked upon now."

The book was published by the Sierra Club and Ballantine in May 1968 under the title The Population Bomb. It would sell about two million copies. After Ehrlich appeared on Johnny Carson's Tonight Show in February 1970, thousands of letters arrived at the network. Lyndon Johnson cited the argument. Walter Cronkite endorsed it in his evening broadcasts. Universities across the United States built population-studies departments around its central claim, which was that no agricultural advance could save the species from imminent and global famine.

This is the story of how, while Ehrlich's book was selling, an Iowan-born wheat breeder in a corrugated-iron shed at a research station outside Mexico City was already spending his fifth year preventing the famine that The Population Bomb said could not be prevented, and how, last week, the climate science community quietly retired its own version of Ehrlich's worst case.

Norman Borlaug had been working on wheat at the Rockefeller Foundation's CIMMYT campus in Mexico since 1944. By the early 1960s he had bred several lines of semi-dwarf wheat that would not collapse under the weight of their own grain when fed nitrogen fertiliser, the way the tall traditional varieties did. The semi-dwarfs could absorb three to four times the fertiliser of a traditional plant and turn the difference into yield. Mexico went from importing half its wheat in the 1940s to exporting it by 1963.

In 1965, against the strenuous opposition of the Indian and Pakistani agriculture ministries, Borlaug arranged the shipment of two hundred and fifty tonnes of his semi-dwarf wheat seed to the Punjab. The 1965 harvest in the two countries was reported as 16.9 million tonnes between them. By 1970 it was 27.4 million tonnes. India stopped importing grain in 1971. Pakistan stopped importing grain in 1968. The famines did not come.

In December 1970 Borlaug travelled to Oslo to receive the Nobel Peace Prize. The committee citation credited him with averting "what otherwise would have been a famine of unprecedented proportions." The Nobel laureate Robert Solow later put it more bluntly: Borlaug had saved more lives than any single human being in history. The widely repeated figure is one billion.

Ehrlich never formally retracted. The book stayed in print. He continued to teach at Stanford until well past his retirement, into his eighties. His subsequent prediction that, by the year 2000, England would no longer exist as a nation, was not formally retracted either. He simply moved on to other concerns, and the population-studies departments quietly began to teach demographic transition instead. The book is still on undergraduate reading lists in 2026, more often as a sociology of forecasting than as a forecast.

The most-cited prediction in The Population Bomb did not become wrong. It had been the worst plausible case under a particular set of assumptions: no Green Revolution, no fertiliser subsidies, no semi-dwarf seed programmes, no Indian or Pakistani policy reform. Once those assumptions broke, the prediction broke with them. Nobody had to declare it wrong. The trajectory simply ran below it.

This is the structural shape of what happened to climate change's worst case last month.

The Intergovernmental Panel on Climate Change has, since its third assessment report in 2001, organised its modelling around a small number of named emissions scenarios. The fourth assessment in 2007 leaned on the SRES A1FI scenario, calibrated to a fossil-intensive twenty-first century with no climate policy. The fifth assessment, in 2014, used the new representative concentration pathway framework, of which RCP8.5 was the highest-emission member. The sixth assessment, in 2021, used the shared socioeconomic pathways: SSP5-8.5 was the highest. RCP8.5 and SSP5-8.5 are technically equivalent to each other, and they were both designed as upper bounds. Their intended use was to ask: if we did nothing, and if coal use continued to expand at twentieth-century rates throughout this century, what would the climate look like in 2100?

For roughly fifteen years they were used by thousands of researchers, journalists, and policymakers as if they were the expected future, not the upper bound. Two-thirds of climate-impact studies between 2010 and 2020 used RCP8.5 or SSP5-8.5 as the headline scenario.

In January 2020, in a Nature commentary titled "Emissions: the 'business-as-usual' story is misleading," the climate scientists Zeke Hausfather and Glen Peters argued that this had become indefensible. The pathway to RCP8.5, they pointed out, required an unprecedented fivefold increase in global coal use over the century, larger than some estimates of recoverable reserves. Coal had already begun to decline. Renewables had fallen in cost faster than any of the scenarios projected. Climate policy, however inadequate, was no longer absent. The trajectory was running below RCP8.5 and showed no sign of bending up to meet it.

The IPCC's sixth assessment in 2021 acknowledged the point in language that was easy to miss. On page one hundred and ten of the first chapter of the working group one report, the IPCC noted that "the likelihood of high emissions scenarios such as RCP8.5 or SSP5-8.5 is considered low." Then, across the same report, it referred back to SSP5-8.5 a further one thousand three hundred and fifty-nine times.

On the seventh of April twenty twenty-six, in a paper titled "The CMIP7 ScenarioMIP design" published in the journal Geoscientific Model Development, an international team led by the Dutch climate modeller Detlef van Vuuren, the architect of the original RCP framework, formally removed RCP8.5, SSP5-8.5, and SSP3-7.0 from the next round of model intercomparison. The paper's exact wording is that "the CMIP6 high emission levels (quantified by SSP5-8.5) have become implausible, based on trends in the costs of renewables, the emergence of climate policy and recent emission trends."

The architects of the worst case have retired their worst case. The other scenarios, the moderate ones in which the world is heading and on which most current policy is based, remain in place.

The arc closes. There has been no press conference. The IPCC has not issued an apology. The headlines that called the apocalypse unavoidable have not been redacted. This is how a scientific worst case dies. Not by retraction. Not even by admission. By a footnote in a model-intercomparison paper, written by the people who built the number, acknowledging that the trajectory has run below it.

Paul Ehrlich is, at the time of writing, ninety-three years old. He still lives in Stanford. He still believes the bomb will go off, just later than he predicted. The Green Revolution averted his version of it. Other revolutions, named and unnamed, are now averting Detlef van Vuuren's.

Borlaug, who saved a billion people, would, one suspects, have approved.