A Century of Trying to Catch a Cold

An Essay

Author’s Note

This essay prosecutes the establishment’s own evidence. Throughout, words like virus, infection, transmission rate, and pathogen are the language of germ theory and of the researchers who ran these experiments. They appear here because the case is built from inside that framework, using its own studies and its own measures against its own conclusions. They are not concessions.

Where the essay states what the record actually shows, the register shifts: people became ill, or they did not, after shared exposures, in conditions the contagion model cannot account for. The reader should keep the two registers distinct. When an experiment speaks of “infecting” a volunteer, that is the claim under examination. When the record shows the volunteer stayed well, that is what happened.

Preface

This essay draws primarily on Daniel Roytas’s Can You Catch a Cold? Untold History and Human Experiments (2024), which documents over a century of human transmission experiments attempting to demonstrate that respiratory illnesses spread from sick people to healthy people. The experimental records cited here come from Roytas’s review of the primary scientific literature.

---

In November 1918, at the height of the deadliest pandemic in recorded history, the United States Navy began a series of human experiments on Deer Island, Boston. Sixty-two healthy sailors volunteered for a sustained attempt to give them Spanish influenza. Researchers sprayed unfiltered mucus from sick patients into their noses and throats. They swabbed their nasal passages with secretions transferred directly from influenza patients. They dripped the sick men’s material into their eyes. They injected filtered mucus under their skin. They injected blood drawn from severely ill patients directly into healthy men.

None of the sixty-two sailors became sick.

The Navy conducted similar experiments simultaneously at Angel Island in San Francisco and Gallups Island in Boston Harbor. Across all three sites, 161 sailors participated in 25 separate experiments over six months. The procedures grew increasingly aggressive. In one experiment, ten healthy men sat at the bedsides of thirty influenza patients at Chelsea Naval Hospital. Each volunteer spoke face-to-face with a sick man for two to three minutes. The patient breathed on him five times, then coughed directly into his face five times while the healthy man inhaled. Each volunteer repeated this with ten different patients. Total exposure time: thirty to fifty minutes of direct facial contact with severely ill men during the peak of a pandemic that would kill between fifty and one hundred million people worldwide.

Not one of the ten volunteers developed influenza.

The final tally across all three islands: three men developed any illness at all. Two of those were classed as influenza. One showed questionable symptoms that may or may not have constituted illness. Two cases in 161 men is an infection rate of 1.2 percent. The experiments were conducted under the authority of the Surgeon General, with consultation from professors at leading universities, and staffed by more than fifty high-ranking naval officers and scientists from multiple military divisions. They represent arguably the most comprehensive, well-resourced transmission experiments ever conducted in the history of medicine.

They failed to demonstrate contagion.

This work stays free because paid subscribers make it possible. They get the full book library, the Deep Dive Audio Library, and the Questions for Your Doctor, Before You Consent, and Package Insert series. No grants, no gatekeepers — your subscription is what keeps it that way.

The Military Campaign Against Influenza

The U.S. military took the Spanish flu seriously. By the summer of 1918, more than 43,000 American soldiers had died, not from combat, but from pandemic influenza. This figure represented eighty percent of total American combat deaths for the entire war. Up to forty percent of Army and Navy personnel were incapacitated by illness at some point during the pandemic. For the military, understanding how influenza spread was a question of survival.

The experiments at Deer Island began in November 1918. The volunteers ranged in age from fifteen to thirty-four, all in excellent physical condition. Of the sixty-two men, thirty-nine reported never having been ill with influenza at any point in their lives. The researchers drew sick donors from the Deer Island Naval Hospital and mainland military hospitals, selecting patients with active, severe cases of Spanish influenza.

The first experiment inoculated six healthy sailors with a pure culture of Pfeiffer’s bacillus, the organism many scientists at the time suspected caused influenza. All six men remained well.

The second experiment sprayed unfiltered mucus secretions from sixteen influenza patients into the nasal passages and throats of twenty healthy men. Some of the mucus was swallowed. None of the men fell ill.

The third experiment instilled unfiltered mucus from four influenza patients into the eyes and nasal cavities of ten healthy sailors. All remained completely well.

The fourth experiment swabbed the nasal passages and throats of nineteen healthy men with mucus transferred directly from ten influenza patients. None were adversely affected.

The fifth experiment injected filtered mucus from three influenza patients subcutaneously into ten healthy sailors. Every man remained well.

The sixth experiment drew blood from five sailors suffering from pandemic influenza and immediately injected it subcutaneously into ten healthy men. Apart from localized muscle soreness at the injection site, none became ill.

The seventh experiment, the bedside exposure described above, produced no cases of influenza in any of the ten healthy participants despite each man’s prolonged, face-to-face contact with ten coughing patients.

The eighth experiment sprayed one billion Pfeiffer’s bacillus bacteria, cultured from thirteen influenza cases, into the noses and throats of nineteen healthy men. None developed influenza.

Across eight experiments on Deer Island involving sixty-two men subjected to every conceivable mode of exposure, the transmission rate was zero.

The Angel Island experiments ran concurrently in San Francisco. Fifty sailors from the naval training station at Yerba Buena participated after being quarantined for one month prior. Their ages ranged from eighteen to twenty-three. Eight separate experiments exposed them to filtered and unfiltered mucus secretions, lung fluid cultures, nasal washes mixed with coughed-up material, eye drops of mucus, subcutaneous injections of mucus, and intramuscular injections of patients’ blood.

None of the fifty volunteers developed influenza.

The Gallups Island experiments began in February 1919, after the pandemic’s peak had passed but while cases remained active. Forty-nine volunteers participated in nine experiments. Researchers sprayed bacterial cultures into nasal passages, instilled patients’ mucus into throats, swabbed nasal cavities with secretions from multiple patients, injected blood subcutaneously, and exposed healthy men directly to the coughs of sick patients.

Two men developed influenza. One developed an influenza-like illness of uncertain origin.

The scientists who conducted these experiments were confounded by the results. They offered no explanation for why so few of their participants developed influenza when exposed to material from patients suffering from the deadliest pandemic on record. They could not identify the causative agent. They could not describe the mode of transmission. Despite their most diligent attempts, they found it all but impossible to infect healthy men with influenza.

The Accumulating Failures

The Navy experiments were neither the first nor the last attempts to demonstrate influenza transmission. They belong to a much larger body of failed experiments spanning decades.

Shortly after the Spanish flu pandemic began, John Nuzum and colleagues conducted three experiments exposing healthy volunteers to mucus secretions and ground-up lung tissue excised from influenza patients. Of seven participants, none developed the disease. Hugo Selter attempted to infect himself and his assistant with the nasopharyngeal washings of five influenza patients. Neither man became ill. René Dujarric injected himself with the blood of four influenza patients. He remained healthy.

Charles Nicolle and Charles Le Bailly conducted a particularly elaborate series. They collected mucus from influenza patients and dropped it into the eyes and nasal passages of a monkey. When the monkey developed a fever, they drew its blood and injected it into two healthy men, one subcutaneously, one intravenously. Both men remained well. The researchers then injected the original human mucus directly into two other volunteers. One developed mild flu-like symptoms. This mildly symptomatic man then shared living quarters with several healthy individuals. Not a single case of transmission occurred.

Ferdinando Michelli and Giuseppe Sata injected filtered mucus and unfiltered blood into eighteen healthy participants. None developed influenza. Frank Schofield conducted four experiments exposing eight healthy people to mucus and blood from influenza patients. All attempts failed, with one exception: when Schofield injected himself and two research assistants with unfiltered blood, he and one assistant developed an influenza-like illness. The clear risk of experimenter bias in a self-inoculation study, combined with the numerous substances present in unfiltered blood, makes this result difficult to interpret.

Frederick Lister and Edward Taylor shipped eleven healthy men to a remote island five hundred miles from civilization, a place untouched by the Spanish flu, and inoculated them with filtered mucus from influenza patients. All remained well. They inoculated nine other men with pure cultures of Pfeiffer’s bacillus; one developed influenza, an odd finding since Pfeiffer’s bacillus is definitively not the cause of influenza. They sprayed five more men with unfiltered mucus; two developed flu-like symptoms. Overall, three of twenty-five participants (twelve percent) showed any symptoms, and one of those cases is immediately disqualified by its bacterial origin.

Harry Wahl, George White, and Harold Lyall ground up lung tissue from a deceased influenza patient, centrifuged it, and sprayed the supernatant into the nostrils of two healthy men. Both remained well. They inoculated pure cultures of Pfeiffer’s bacillus into the respiratory tracts of seven men. All remained completely healthy. Arthur Bloomfield attempted to infect fourteen healthy men. Not a single one developed the flu.

Anna Williams, Mary Nevin, and Caroline Gurley inoculated forty-five healthy participants with bodily secretions from cold and flu patients. None became ill. Robert Robertson and Robert Groves exposed one hundred healthy volunteers to filtered mucus secretions of the common cold. No one fell ill. Oscar Costa-Mandry, Pablo Morales-Otero, and Jenaro Suarez attempted to infect eighteen healthy men with unfiltered and filtered influenza mucus. All attempts failed.

By the time the Spanish flu pandemic ended in 1920, scientists had conducted dozens of human experiments. They had made no progress toward understanding the cause of the disease. The field of medicine was no closer to understanding influenza than before the pandemic began.

The Common Cold Research Unit

The pattern of failure continued for decades.

In 1946, the British government established the Common Cold Research Unit in Salisbury, England, a dedicated facility tasked with identifying the cause of the common cold and finding a cure. The unit consisted of twelve flats housing up to thirty volunteers at a time for ten-day research trials. Participants received all-expenses-paid stays, reimbursed at £1.25 per day. Upon arrival, volunteers were quarantined from each other and inoculated with either alleged cold virus material or placebo in single- or double-blind fashion.

In its early years, virologists at the CCRU attempted to infect twenty different animal species with common colds: rabbits, guinea pigs, rats, mice, cotton rats, hamsters, voles, squirrels, ferrets, kittens, pigs, hedgehogs, and several species of monkey including baboons, sooty mangabeys, brown capuchins, red patas, and green monkeys. Their attempts were entirely unsuccessful.

They then turned to human experiments.

In one of the more rigorous trials, researchers inoculated healthy participants with a solution grown in chicken embryos, said to contain cold virus. Before symptoms developed, they exposed inoculated participants to pairs of healthy people for ten hours to test whether colds could be passed on during the incubation period. None of the eight healthy participants became sick. A few days later, the four inoculated participants with the worst symptoms were exposed to eleven healthy people. One developed a cold. When nine healthy participants were exposed to people with experimentally induced colds, one developed mild symptoms. When five healthy people were exposed to a person with a naturally acquired cold, they were unaffected.

The virologists were perplexed. They wondered if people living in isolated communities might be more susceptible. To test this, they sent twelve healthy volunteers to a remote island in the Hebrides where no one had lived for over a decade. The volunteers had supplies to last three months and remained completely isolated for ten weeks. Then six participants with experimentally induced colds were sent to the island.

The first trial tested whether colds could spread via fomites, contaminated surfaces. One group of healthy volunteers vacated their dwelling while the six sick participants smeared their nasal secretions over items and surfaces inside. The sick participants then left before the healthy ones returned. Four healthy participants were deliberately exposed to this contaminated environment. None became unwell.

The second trial tested aerosol and droplet transmission. The sick participants interacted in a room with four healthy participants, coughing and sneezing on them for three hours. None of the healthy participants became sick. A new group with experimentally induced colds spent six hours with healthy participants. All remained well. In an extended exposure, healthy people lived with infected participants for four days. None fell ill.

The researchers then located a farmer on a nearby island who had developed a natural cold five days earlier. They brought him to interact with one group of healthy participants for two hours. Three of four developed colds. He then spent nearly four hours with a second group. None of them developed a cold.

Across its early human trials using filtered and unfiltered nasal secretions, the unit inoculated roughly 231 subjects and confirmed about 137 cases of cold, a case rate near 59 percent. It dwarfed most other studies, and yet the researchers expressed disappointment, doubt, and caution. They regarded the result as a “will-o’-the-wisp,” something impossible to grasp, and admitted their techniques were uncertain and difficult to interpret. From a germ perspective they considered even this rate a failure. They disliked using human subjects because of how resistant the subjects were to catching colds, despite enormous doses delivered by unnatural means. In one trial, participants were inoculated with the combined nasal washings of twenty-six sick people.

Between 1952 and 1953, 292 participants were inoculated with cell culture material allegedly containing cold viruses. Only fifteen (five percent) developed colds. The unit was on the verge of being disbanded.

A new director, David Tyrrell, was appointed in 1957 in a last-ditch effort to salvage the research program. Under his management, the CCRU began using tissue and cell culture methods to “isolate” cold viruses. By the time the unit closed in 1989, more than 1,000 papers had been published and nearly 20,000 individuals had participated in experiments. Of those inoculated with alleged cold viruses, between one-fifth and one-third developed colds.

That “success rate” came under the most favourable conditions imaginable. The CCRU was delivering concentrated forms of allegedly highly infectious particles by direct inoculation, conditions far more favourable to transmission than any natural encounter. Under those artificial circumstances, a highly contagious pathogen should have produced near-universal illness. A one-fifth to one-third rate instead describes something that barely produces illness even when researchers do everything they can to force it.

The CCRU’s methods also created systematic biases that likely inflated their results. The unit advertised research opportunities as all-expenses-paid holidays and reimbursed participants for their time. This attracted a non-random sample: people who might be poor and desperate, stressed and in need of respite, or hypochondriacs seeking validation. The incentives also encouraged repeat participation. Many volunteers returned for second, third, and fourth visits; some came up to nine times. Testing the same people repeatedly could reinforce learned responses, conditioning volunteers to expect and manifest symptoms based on previous experiences.

Participants kept records of their own symptoms, which the CCRU admitted “naturally influences the doctor’s decision” to count them as having a cold. Self-monitoring combined with self-selection and repeated exposure creates conditions for inflated case counts. The phenomenon of “leisure sickness,” where people fall ill with colds and flu shortly after going on holiday, may also have contributed. The CCRU marketed participation as a ten-day vacation, potentially priming volunteers to experience this well-documented effect.

Even with these biases working in favor of positive results, the CCRU could not reliably transmit colds. Staff virologists admitted that one of their biggest challenges was simply giving people colds. For a supposedly common and highly contagious illness, this difficulty is remarkable.

The Methodological Paradox

Across this body of research, transmission rates and methodological rigor run in opposite directions. The studies with the highest infection rates have the most serious methodological flaws. The studies with the strongest methods show the lowest transmission.

Professor Tamotsu Yamanouchi’s Japanese research team published results in June 1919 that stand out dramatically from the pattern. They combined lung fluid from forty-three influenza patients into a single concoction mixed with Ringer’s solution and inoculated it into the nasal passages and throats of twenty-four healthy recipients. Eighteen (seventy-five percent) developed influenza symptoms. They injected some of the mucus subcutaneously into eight other people; seven developed influenza. They injected influenza patient blood into six healthy subjects; all six developed influenza.

These rates dwarf every other study, and the methods account for the difference.

Yamanouchi’s team used no control groups. Ringer’s solution, the isotonic fluid they mixed with mucus samples, produces adverse effects on its own, including coughing, sneezing, nasal congestion, and difficulty breathing. Without a control group receiving Ringer’s solution alone, or lung fluid from healthy people, the researchers could not distinguish symptoms caused by any alleged virus from symptoms caused by the solution.

They used no random sampling. All volunteers were friends and colleagues of the researchers, a non-representative sample that limits generalizability and introduces systematic bias.

They used no blinding. Participants could easily determine or be told what the experiment was testing, creating conditions for nocebo effects. The desire to reciprocate or contribute to the research of friends and colleagues could influence symptom reporting.

They never isolated any virus. The solution injected contained the combined bodily secretions of forty-three sick people, a scenario that would never occur in nature.

No other study replicated Yamanouchi’s results. The most compelling evidence for contagious influenza came from the least reliable methods.

Paul Schmidt’s experiments in 1920 revealed a related problem. In one study, he inoculated 196 healthy participants with filtered mucus drawn from sixteen people with common colds. Twenty-one developed colds and three developed influenza. In a second study, he gave eighty-four men the filtered mucus of twelve influenza patients; five developed influenza and four developed colds. His final study removed the sick material altogether. He inoculated forty-three men with physiological saline alone, containing nothing from any patient. Eight of them, 18.6 percent, developed common colds. That is a higher rate than either group given the actual secretions of sick people, where colds appeared in under eleven percent.

This creates a fundamental problem for every transmission experiment that mixes bodily fluids with other substances. If plain salt water produces colds at a higher rate than the mucus of the sick, the researchers cannot say which substance produced the illness. Any study without a proper control group, and most have none, cannot attribute symptoms to a transmitted agent.

The absence of adequate controls pervades this research. Without a control group, scientists have no way of knowing which variable produces observed symptoms. The mucus secretions of sick people contain far more than alleged viruses. They contain inflammatory mediators, cellular debris, proteins, and countless other substances. Any of these could produce respiratory symptoms when introduced into healthy nasal passages.

In 1922, Dr. Victor Vaughan, Professor of Hygiene and Chairman of Medical Sciences for the National Research Council, along with Dr. Henry Vaughan, Commissioner of Health for Detroit, and epidemiologist Dr. George Palmer, published a medical textbook addressing this exact issue. They explained that symptoms resulting from sick-to-healthy inoculation of filtrates are caused by reactions to non-viral proteins suspended in the mucus. When these proteins are introduced into nasal passages, they cause local sensitization and inflammation. The authors concluded that introducing any substance containing proteins into the upper respiratory passages can cause reactions including common colds, epidemic coughs, and acute coryzas.

Experiments have demonstrated this principle. Healthy volunteers inoculated with normal chicken embryo fluid, containing no alleged virus, developed common cold symptoms including nasal irritation, sneezing, nasal obstruction, swelling of the nasal mucosa, headache, and elevated body temperature. People developed blocked noses, sneezing, and headaches after being inoculated with sterile broth. They became self-convinced of having developed a cold despite never having been exposed to any pathogen.

The nocebo effect compounds these problems. Participants attending a facility called “The Common Cold Research Unit,” subjected to elaborate procedures by official-looking researchers using high-tech equipment, have every reason to expect they might develop cold symptoms. The more impressive an intervention appears, the greater its placebo or nocebo effect. Participants were not blind to the fact that they were being experimented on and that there was a good chance of receiving cold germs. This knowledge produces effects that move in the same direction as the researchers’ hypothesis, systematically biasing results toward positive findings.

True blinding is nearly impossible in these experiments. Even when researchers conceal whether a participant receives active or control intervention, contextual cues remain abundant. Any positive result must therefore exceed what nocebo effects, allergic reactions, and non-viral irritants would produce on their own. Without rigorous controls, no study can claim to have demonstrated transmission as opposed to some combination of these alternative mechanisms.

The Counterarguments

Two explanations are usually offered for these failures.

The first appeals to strong immunity. Perhaps the sailors and volunteers had robust immune systems that protected them from infection.

This explanation fails on mathematical grounds. The Spanish flu is said to have infected roughly 500 million people out of a global population of 1.8 billion, around twenty-eight percent of humanity. The rate in the Navy experiments was 1.2 percent, a discrepancy of more than twentyfold. That twenty-eight percent accumulated across multiple waves over two years; even halved to fourteen percent for a single wave, or reduced to nine percent, it leaves a gap far too large for immunity to close.

The volunteer population makes the discrepancy more striking, not less. The sailors were young, healthy men, precisely the demographic that the Spanish flu killed at the highest rates. Males between twenty and forty accounted for roughly half of all Spanish flu deaths, an inversion of the typical pattern where children and the elderly are most vulnerable. Young healthy men should have shown substantially higher infection rates than the general population. Instead, they showed dramatically lower rates.

If strong immunity explained the experimental results, the same immunity should have protected this demographic in the general population. It did not. Young healthy men died in unprecedented numbers during the pandemic while proving nearly impossible to infect under controlled conditions.

The second explanation appeals to weakened pathogenicity. Perhaps the experiments exposed participants to a less virulent strain late in the pandemic.

This explanation also fails. The donors were genuinely ill, some fatally so, and several of the doctors running the experiments died. Whatever was killing people was doing so during the experiments, using their secretions as the source material. The Deer Island and Angel Island studies ran in November 1918, at the pandemic’s deadliest wave, not in some attenuated aftermath. And the volunteers were drawn from the very group the pandemic killed most. If susceptibility tracked the general population, they should have fallen ill at higher rates than average, not lower. They fell ill at 1.2 percent.

The Studies That “Succeeded”

Several experiments do claim successful transmission. None holds up.

Gwaltney’s 1970s experiments produced mixed results. In the first trial, one person with a cold sat at a table talking loudly, singing, coughing, and sneezing for fifteen minutes with groups of healthy participants. Twelve healthy people were exposed to six different sick donors. None developed a cold.

In the second trial, ten healthy participants were housed in a large room for three days with six people suffering from colds. Wire mesh prevented physical contact but allowed air circulation. The sick donors coughed and sneezed freely. No fresh air was introduced, maximizing exposure to “infected air.” Not a single healthy volunteer developed a cold.

In the third trial, six sick donors blew their noses into their hands. Fifteen healthy volunteers touched the donors’ hands for ten seconds, then touched their own eyes and noses. Nine of fifteen developed cold symptoms.

The overall rate across all three experiments: nine of thirty-seven participants (twenty-four percent). But this result emerged entirely from the third experiment. The first two showed zero transmission via droplets, aerosols, or airborne particles. Only direct hand-to-hand-to-mucous-membrane contact produced any illness.

Further problems emerge from Gwaltney’s diagnostic criteria. He scored symptoms on a scale from zero to three and considered participants to have caught a cold if they achieved a minimum score of six points. This threshold was arbitrarily low; the scoring method he adapted from another experiment required a minimum score of fourteen. One “positive cold” had a symptom score of seven, consisting of scant nasal secretions and a single sneeze over three days. Their own testing found no virus in the mucus. This participant was categorized as having a cold.

D’Alessio’s 1984 experiments fared no better for the contagion hypothesis. In two trials, groups of healthy participants sat at tables with sick people, played cards, and conversed freely. No one developed a cold. In another trial, five pairs of healthy people spent twelve hours per day for three consecutive days in a room with two sick donors. One of ten developed suggestive symptoms. In two trials where healthy people kissed sick people for over a minute, one of sixteen developed mild symptoms.

A 2020 experiment attempted sick-to-well transmission of influenza with 127 participants. Fifty-two healthy participants were inoculated intranasally with cell culture fluid allegedly containing influenza virus. Forty-two (eighty-one percent) developed flu-like symptoms, an apparently strong result. But no adequate controls were used. Whether symptoms resulted from any virus, cell culture fluid components, or nocebo effect cannot be determined.

The forty-two symptomatic participants were then exposed to two groups of healthy people for four days under household-like conditions. One group of forty wore face shields and sanitized their hands every fifteen minutes. One group of thirty-five used no protective measures. Of seventy-five healthy people exposed to sick people, one (1.3 percent) developed symptoms. The lab-induced illness, whatever its cause, did not readily transmit.

The Gap Between Acceptance and Demonstration

A 2003 literature review searching for evidence of human-to-human influenza transmission could not find a single published study delineating contagion routes. A similar review conducted years later found no conclusive evidence demonstrating transmission routes and concluded that sick-to-well transmission is not how influenza spreads. Other researchers have found that the natural spread of respiratory illnesses is “incredibly difficult” to replicate under experimental conditions.

Systematic reviews conclude that airborne transmission is unlikely to be significant in the spread of respiratory illnesses, and that there is no evidence showing disease transmission through large respiratory droplets. Studies have highlighted the lack of evidence supporting transmission of rhinovirus and coronavirus via fomites. Reviews have noted an absence of scientific data showing aerosol transmission of SARS-CoV-1 and SARS-CoV-2. One research group stated explicitly that there is no direct evidence for transmission of SARS-CoV-2 via any route: fomites, direct contact, droplets, or aerosols.

The experiments fall into two groups. Most produced no illness at all, a body of null results made larger still by the publication bias that buries negative findings. The minority that produced illness did so inconsistently, under methods with no control groups, no blinding, no isolated variable, diagnostic thresholds low enough to count a single sneeze, and contradictory outcomes across routes that should have been equivalent.

Over two hundred human transmission experiments spanning more than a century have attempted to demonstrate what is taken as self-evident: that respiratory illnesses spread from sick people to healthy people through close contact. The overwhelming majority failed. The minority that succeeded did so inconsistently and with methodological shortcomings that prevent confident conclusions. The most rigorous experiments, the Navy studies during the Spanish flu pandemic, showed transmission rates below two percent despite exposure methods far more aggressive than any natural encounter.

When these experiments first emerged, experts doubted that invisible filter-passing agents caused Spanish and seasonal influenza, given the consistent failures and the weakness of the few positive results. One might assume scientists have since conducted improved human-to-human contagion experiments that resolved those doubts. They have not. The experimental record has not substantially changed.

Two forces explain why this record has left so little mark. The first is publication bias: studies showing no transmission are far less likely to be printed than studies claiming a positive result, so over decades the literature fills with apparent confirmations while the null results vanish into file drawers. The second is professional incentive. Whole departments, institutes, and funding streams rest on the assumption that respiratory disease spreads person to person via identifiable agents. To question that is to question livelihoods and institutions, not merely a conclusion. The skepticism that met the Navy experiments did not survive, and it was not retired by better experiments, because better experiments were never run. It was worn away by repetition.

What Was Killing Them, If Not Contagion

If the disease was not passing from person to person, the deaths of fifty to one hundred million people need another explanation. The records point away from contagion and toward exposure. The pandemic’s deadliest months coincided with the heaviest use of phosgene and chlorine gas in human history, and pathologists of the period could not tell the lungs of gas casualties apart from those of “flu” victims. Aspirin was dispensed at doses now known to cause fatal pulmonary toxicity. Four years of war had left whole populations malnourished, exhausted, and subjected to mass inoculation campaigns using preparations laced with heavy metals and solvents. These are shared exposures, concentrated in the same camps and cities, not an agent travelling from one body to the next.

One hundred sixty-one sailors were exposed to Spanish influenza by every route their doctors could devise: sprays into the nose and throat, swabs, injections beneath the skin, transfusions of patients’ blood, and face-to-face contact with men coughing directly into their open mouths. Ninety-eight percent stayed well, during the deadliest pandemic on record, drawn from the very group it killed most. The Common Cold Research Unit spent four decades and twenty thousand volunteers trying to pass a cold from one person to another and could not reliably do it. The studies are published, and their numbers have never been disputed. They have never been allowed to mean what they say.

---

Explain It To A 6 Year Old

Imagine a group of doctors who really, really wanted to prove that you can catch a cold from another person. So they tried. They took the snot from very sick people and sprayed it up healthy people’s noses. They wiped it in their eyes. They had sick people cough right into their open mouths, over and over.

Almost nobody got sick.

They did this with sailors during the worst sickness the world had ever seen, and ninety-eight out of every hundred stayed perfectly well. Years later, a whole building full of scientists in England spent forty years trying to give people colds on purpose. They mostly couldn’t do that either.

So if coughs and sneezes and snot don’t really pass sickness from one person to another, why do people in the same house sometimes get sick around the same time? Because they are sharing the same things: the same bad air, the same food, the same worries, the same cold and tiring days. They aren’t catching it from each other. They are all bumping into the same problem at the same time.

When huge numbers of people got very sick long ago, it wasn’t a tiny bug hopping from person to person. They were breathing poison left over from a war, being given too much medicine, and were tired and hungry. Their bodies were dealing with real things you could point to, not an invisible bug sneaking around.

---

References

Roytas, Daniel. Can You Catch a Cold? Untold History and Human Experiments. 2024.

Engelbrecht, Torsten, Claus Köhnlein, Samantha Bailey, et al. Virus Mania. 3rd English Edition, 2021. (Context for the 1918 toxic-exposure record: phosgene gas, aspirin toxicity, wartime conditions, and mass inoculation.)