Skip to content

The Unwinnable War on Disease

Pandemics and the ecology of armed conflict

In my book New Pandemics, Old Politics, I tell the history of the modern world in four pandemics. It begins with cholera in nineteenth-century Europe and South Asia and continues with the Great Influenza of 1918–19 and HIV/AIDS in Africa and Western countries. The fourth case is “Pandemic X” a.k.a. “the next big one”—the feared and imagined coming plague against which we have been constructing defenses for the last twenty-five years. Those defenses failed to stop Covid-19. Doubtless the world could have done much better, especially had there not been a dedicated skeptic of science in the White House. But I worry that in choosing to “fight” infectious microbes to “defend” our way of life, we are locking ourselves into a conflict we cannot win.

I use the “war on disease” as an organizing principle to examine two centuries of scientific thinking about epidemic diseases and the politics that have shaped the responses to each pandemic. As the title of the book indicates, each pandemic is new—indeed its newness is its defining characteristic—while the political and social responses to it are dully predictable. One pattern is that no pandemic is “natural”: each killer pathogen evolves and spreads in an ecology that we ourselves have engineered. For cholera and HIV/AIDS, it was European conquest and the reorganization of colonially subjugated societies for the profit of their imperial masters. For influenza, it was the organization of total war.

The excerpt below is from the chapter on influenza. It illustrates the irony of approaching the greatest pandemic of the twentieth century through the lens of the “war on disease,” because it was in fact the war itself—and specifically the advances in military hygiene and infectious disease control—that created a unique ecology in which this exceptionally virulent strain of influenza could evolve and spread.


Franz Kafka contracted influenza in October 1918. He was living in Prague as a subject of the Austro-Hungarian empire, ruled by the Hapsburgs, heirs to the Holy Roman Empire. He fell desperately ill while a long-awaited revolution was played out in the streets below his family’s apartment and rose from his sickbed five weeks later as František Kafka, citizen of the newly declared Republic of Czechoslovakia. In the words of his biographer Reiner Stach, it was “eerie but a bit comical.”

The influenza pandemic is the deepest mystery in the history of disease.

As the denouement of the Great War swept through Central Europe, the influenza pandemic “slipped to the third page of many daily newspapers,” Stach writes. On the first day of Kafka’s illness, troops cordoned off the Altstäder Ring, where his family apartment was located. The next day they melted away and the protesters took over the streets. “Just when there were finally good reasons to prohibit large gatherings, the public space had slipped out of government control,” Stach observes. “Demonstrations and national rallies had become hot spots where history was being made—and people were not covering their mouths with handkerchiefs.” Kafka had long imagined this day; he had written about what the end of the war and the end of empire would mean for his own career as a writer. But as the political drama unfolded under his window, he was deep in a flu-induced fever, from which—already weakened by tuberculosis—he never fully recovered.

Stranger still than Kafka’s own story is the way in which the locomotive of world history metamorphosed—for that instant—from the god of war into the tiniest quasi-living thing, an assortment of RNA that is the shapeshifting influenza virus. Writing the story of influenza demands Kafka’s ability to see what is alien in the familiar and what is normal in the strange. The influenza pandemic is the deepest mystery in the history of disease.


The Great Influenza of 1918–19 is the totemic pandemic of modern times. It is the biggest-ever mass mortality episode of the last two hundred years, regardless of what metric is used—total numbers or proportion of population. As historical demographers have become more skilled in their use of statistical methods, the estimates have crept up from 21 million dead at the time to today’s best figure of somewhere between 60 million and 100 million. This wide span of estimates is a commentary on influenza’s disguise as an everyday ailment and its propensity to kill indirectly, through making other illnesses such as tuberculosis or heart disease more lethal. The toll is at least three times more than those killed in the Great War and commensurate with the combined total of deaths from violence and hunger during World War Two.

The 1918 strain of influenza possessed an unprecedented combination of transmissibility and virulence. Most pathogens score high on one of these characteristics and low on the other. For instance, an easily transmitted airborne virus that needs its host to be mobile tends not to cause rapid, often lethal symptoms, while a nosocomial (hospital-based) bacterium usually is the reverse. The influenza virus is prone to mutation: it has an annual lottery of genetic reassortment that reshuffles its traits, normally without major changes.

Suddenly, in the first half of 1918, a variant was born supremely well-aligned for velocity of transmission, speed of disease progression to lethality, and selecting young adults, especially men. It raced around the world and then vanished. No subsequent strain has come close to matching its power. The pandemics of 1928, 1957, and 1968 each killed more than a million people, but the special qualities of the 1918 virus remained a puzzle until virological science was able to unpick its lock. This was possible only with scientific advances—especially genetic sequencing—at the turn of millennium.

Why did influenza become four orders of magnitude more dangerous, at that time and in those places?

A team led by Terrence Tumpey at the U.S. Centers for Disease Control working under strict biosecurity protocols was able to obtain samples of the virus from victims—one whose lungs were preserved at a military hospital, the other recovered from a body frozen in the Arctic permafrost. The team worked at night when the laboratories were otherwise deserted. They succeeded and reconstructed a live virus. It was astonishingly virulent: injected into mice, it replicated at a rate that after four days produced a viral load 39,000 times higher than other influenza strains.

This was a remarkable achievement. Tumpey considered it the virological equivalent of the moon landing. But it didn’t solve the origins puzzle. His team concluded that the ultimate source of the 1918 virus was an avian virus, but the pathway to the final pandemic form could not be determined—for example it might have passed through an intermediate mammalian host. Another crucial finding was that there was no single element that made the virus so deadly, but rather, “the constellation of all eight genes together make an exceptionally virulent virus.” If they had been able to isolate strains from the immediately prior influenza viruses circulating, they would have been able to say more. That wasn’t possible from the fragments of viral archeology available. And more clues to the genealogy wouldn’t have answered the bigger question of why: Why did influenza become four orders of magnitude more dangerous, at that time and in those places? To address that question, we need to apply a different disciplinary framework: Darwinian natural selection in the context of man-made human ecology.


One way of defining a pandemic would be that it is the emergence of a new microbe, so well adapted to the human body and society, that we must adapt to it. The outcome of a pandemic is the co-evolution, or hybridization, of us and the pathogen. The natural selection of both is shaped by medical technologies such as vaccines and social technologies ranging from airplanes and schools to plumbing systems and air conditioning. In 1918, that list would have begun: war and mobilization for war.

The quarter century from Robert Koch’s isolation of the cholera bacillus to the outbreak of influenza was the heroic age of microbial medicine. The popular plot for the conquest of infectious disease was already written: scientists would identify the invisible enemy, develop a cure or a vaccine, and in the meantime, the government would apply, with military efficiency, methods of transmission control to contain the outbreak. These methods were applied to control cholera in Europe and (selectively) Asia, and yellow fever in the Americas, with impressive results. Some of the biggest health constraints on the white man’s imperial project and on the organization of mass warfare were lifted.

After August 1914, obsessed with pouring more and more men and materiel into the Western Front—driven by what the critics of escalatory militarism called the “rage of numbers”—generals and their civilian masters created an ecosystem without historical precedent. All the environmental-evolutionary components were put in place for a super-transmissible, hyper-virulent airborne pathogen to evolve, find itself perfectly fitted to this man-made ecology, and explode across the world. It thrived in training camps, trains, and troopships, not least because the best military medicine had suppressed the traditional campaign diseases such as typhus and cholera while army medical corps provided ambulances and hospitals for the wounded and sick. In a revealing irony, the so-called victories of the so-called war on disease were what opened the door for the worst pandemic in history.

When the Great War began, Britain’s chief medical officer addressed the troops advising them that they needed to defend against bacilli as well as bullets, though they should be alert but unafraid of the former. His German counterparts commissioned a systematic study of the “war pestilences” that would imperil troops in every territory through which they advanced. Chief among their fears was typhus, which has a good claim for having decided the outcomes of the Balkan front in 1914–15 and the Eastern Front in 1916–17, when it selectively ravaged the armies of the belligerents at crucial moments, though it was well-controlled on the Western Front.

When the United States entered the war in 1917, U.S. Surgeon General William Gorgas—promoted to that position on account of his record in the yellow fever campaigns in Cuba and the Panama Canal—promised that the war with Germany would be the first in which fewer soldiers died of disease than in combat. Speaking to American Magazine in March 1918, he stated: “At this stage of the war many commanders report that their soldiers are in better health, even in the trenches, than our civilian population is here at home. There are fewer colds on the battle line in France than there are on Broadway. And the same thing is true of more serious troubles.” That did not turn out to be the case for influenza.

Medical officers at the time pointed to the ways in which the war and the mobilization for war, notably Atlantic troop transports and supply ships, facilitated the faster and wider spread of the virus. What wasn’t understood then was the possibility that the virus itself might be specially adapted to the conditions of war. The case for this was first put forward in 1994 by Paul Ewald, theorist of viral evolution. He argued that the war machine was an ecology in which the virus no longer faced a trade-off between transmissibility and virulence. Industrial war was also a relentless choreography of circulating millions of men from one overcrowded setting to the next. Ewald writes:

The environmental conditions associated with the trench warfare of World War I could hardly have been more favorable for the evolution of extreme virulence of airborne pathogens like influenza. Soldiers in the trenches were grouped so closely that even immobile infecteds could transmit pathogens. When a soldier was too sick to fight, he was typically removed from his trenchmates. But by that time trenchmates often would have been infected because rates of shedding are highest at the onset of illness, which typically occurs two to three days after exposure. The sick individuals were generally moved between a succession of crowded rooms by a succession of crowded vehicles. Severely ill soldiers were transported along with the wounded to field hospitals, where they were usually laid on blanket-covered straw inside tents.

Army doctors quickly recognized how transmission was occurring—it was organized. Front line troops were rotated every two weeks in crowded trains. Packing American troops into overcrowded ships for transport across the Atlantic became a scandal among the small circle of military physicians and generals who were permitted to know the facts. After the war, influenza was added to the ranks of “war diseases” and military physicians feared that it would inevitably accompany future mass mobilization.

The key point here is that the massive, coordinated, routinized, sustained confinement and rotation of young men, including a consistent pattern of exposing the uninfected to the infected—including those rendered immobile by infection—created a unique ecology in which an airborne pathogen could become as virulent as a hospital-based infection. Influenza viruses are prone to genetic change, but only in this special environment would natural selection make a mutation fit for exponential growth to pandemic scale. The evolutionary account explains the much-noted and puzzling anomaly that 1918 influenza selectively struck young men.


British army physicians reported outbreaks of severe respiratory infections at the military camp at Étaples in northern France in 1916, which had high mortality (40 percent) but very little person-to-person spread. This may well have been the forerunner of the pandemic variant—at least some doctors thought so at the time. Then, in April 1918, a sharp rise in influenza cases was reported among German soldiers. Two months later, as Germany massed troops for its final offensive on the Western Front, a virulent plague struck. General Erich von Ludendorff had redeployed millions of reinforcements from the Eastern Front, where the revolutionary government in Russia had agreed to peace terms. He was rushing to strike a fatal blow on the British and French lines before the American troops arrived en masse.

The progeny of 1918 pandemic influenza are still with us, reaping their seasonal toll and occasionally recombining to generate a pandemic.

In war, as in epidemic control, timing is everything. Just as the concentration of German troops and armor reached its peak, influenza mortality rates shot up. A graph published by the German Bureau of Sanitation shows influenza deaths in the army during the war. It has seasonal waves peaking each late winter, and in June 1918 the line literally goes off the charts. From August 1917 until May 1918, mortality from influenza among the troops was between 3 and 4 percent of the caseload, about where it had been since the beginning of the war. In June, 135,002 soldiers contracted influenza on the Western Front, with a case-mortality rate of 35 percent. In July, 374,000 soldiers on the Western Front fell sick and the mortality rate also went off the charts, passing 100 percent—which must reflect a time-lag between the wave of infections peaking even while the wave of deaths was still increasing. In the Allied trenches, officers noted the puzzling lull in the German offensive, little knowing that the onslaught they were about to face would be from viruses not artillery shells.

Ecological explanations aren’t testable in the same way that other scientific hypotheses are. It’s impossible to rule out origins stories that locate the first cases either in the American Midwest or in southern China. Yet those other accounts don’t explain why 1918 influenza was so much worse. Only the ecological approach helps us understand why this particular strain vanished so quickly after the war’s end, as those special ecological conditions no longer existed.

Except that 1918 influenza didn’t disappear completely. When Tumpey’s team reconstructed the virus, they discovered that every succeeding strain of influenza is a direct descendent. In fact, the last hundred years of influenza can best be understood as a single pandemic, one event lasting more than a century—a troubled, unstable mutual adaptation of host and parasite. Today, French and Belgian farmers still reap unexploded shells in their fields during every spring ploughing season, and some—a dwindling toll as the decades go by—lose their lives or limbs to this iron harvest. In a similar way, we are falling sick and dying every winter from the viral harvest of that Great War, because the progeny of 1918 pandemic influenza are still with us, reaping their seasonal toll and occasionally recombining to generate a pandemic. Among the legacies of the Flanders trenches is that we are still catching the flu that hatched there.