The Playbook: Substitution, Assumption, Persuasion

As the conventional history of virology tells it, virologist John Franklin Enders “isolated” and “proved” the measles “virus” in 1954 in his paper Propagation in Tissue Cultures of Cytopathogenic Agents from Patients with Measles. He did this by developing a culturing method that used throat washings from suspected measles patients—collected by gargling fat-free milk and assumed to contain a “virus.”—which were then added to human and monkey kidney cells along with bovine amniotic fluid, beef embryo extract, horse serum, antibiotics, soybean trypsin inhibitor, and phenol red. After several days of incubation, Enders observed morphological signs of cellular degeneration—what he had earlier termed cytopathic effects (CPE) in 1949—which he attributed to the presence of a replicating “virus.” However, he had seen this same effect in “uninfected” cultures, and it was observed in only five of seven clinically diagnosed measles cases. This contradictory evidence did not deter Enders from proceeding. Based on these cell changes and indirect “immune-serum” reactions, the measles “virus” was declared “isolated,” and the culture method Enders developed became a foundational pillar of modern virology.

What is often omitted from the historical narrative is that Enders did not claim that he had conclusively proven that his “agent” was the measles “virus” in his 1954 paper. Although he stated that he had “considerable indirect evidence” supporting an etiological role, he explicitly acknowledged that two critical experiments still had to be performed before causation could be established—a standard formalized as Koch’s Postulates. These were: (1) experimental production of measles in monkeys and in humans using his tissue culture material, and (2) re-isolation of the same agent from those newly sickened hosts.

Enders wrote:

“Although we have thus already obtained considerable indirect evidence supporting the etiologic role of this group of agents in measles, 2 experiments essential in the establishment of this relationship remain to be carried out. These will consist in the production of measles in the monkey and in man with tissue culture materials after a number of passages in vitro sufficient to eliminate any virus introduced in the original inoculum. The recovery of the virus from the experimental disease in these hosts should then be accomplished.

Conclusion. The findings just summarized support the presumption that this group of agents is composed of representatives of the viral species responsible for measles.“

Despite the caution expressed in his paper, contemporary media quickly transformed Enders’ tentative findings into claims of definitive discovery. In a November 19, 1954 article titled Enders States Measles Virus Now Isolated, The Harvard Crimson reported that Enders had “very suggestive” evidence that he had “isolated and grown the long-sought measles virus.” Enders was quoted as saying, “All evidence points to the conclusion that we have successfully isolated the virus for the first time.” He cited “antibody” reactions as strong support for the view that the “suspected virus is itself measles.” Yet even as these claims circulated publicly, he acknowledged that “final proof” was still lacking, admitting: “But we have not yet completed the final proof of inoculating a susceptible monkey and actually producing the disease.” The article further noted that this “conclusive proof” would be delayed until Enders returned from Stockholm, where he was to receive his Nobel Prize.

The following day, The Key West Citizen echoed the story, with Enders emphasizing that while certain tests suggested the agent might be the causative organism of measles, its role remained unproven until disease could be recreated in an experimental animal:

“He announced Wednesday “very suggestive” evidence of the long-sought isolation of the elusive measles virus—and of possible means for its laboratory growth. The method described uses a “tissue culture” method like that which paved the way for the Salk vaccine for polio.

Enders told an audience of scientists at the National Institutes of Health that he and assistants had found in washings and blood of measles patients “an agent” which has passed certain tests indicating it may be the actual causative organism of measles.

It remains to be proved, he said, that it actually is the measles virus—and such proof must await production of measles in an experimental animal with the newfound agent.”

Enders further noted that if the proof could be established, then it might be possible to develop a skin test or a vaccine. In other words, he explicitly did not view the 1954 findings as definitive.

On December 10, 1954, while reporting on Enders’ Nobel Prize award, The Harvard Crimson reiterated that Enders believed he had isolated the measles “virus,” while again conceding that final proof required recreating the disease in animals—something that still had not been done. The article emphasized that Enders had not yet inoculated monkeys with the cultured material and that confirmation remained pending, though he expressed confidence that disease would result:

“Nor have Enders and Weller yet reached the limits on research with tissue culture methods. Ender now believes he has isolated the measles virus–by using the same basic process.

“It’s not a very serious disease although though it does produce deaths and harmful secondary effects in parts of the world like Yugoslavia,” Enders notes. He is not the least apologetic about taking up a new project of less significance than that of isolating the polio virus. “After all,” he notes, “measles are an abominable nuisance.”

Final proof of measles virus growth must await Enders’ return from Sweden, however. For the 57-year-old research scientist has not yet innoculated monkeys with the test-tube grown measles virus.

Enders has no doubt that animals, if exposed, or innoculated will come down with measles.”

This is especially important given the methodological limitations of the study itself. By Enders’ own admission, both in his 1954 paper and in public statements, critical links in the logical chain of causation were absent. In formal terms, these criteria—summarized as Koch’s Postulates—require that the suspected microorganism be consistently found in all cases of disease, isolated independently of host material in pure culture, shown to reproduce the disease when introduced into a healthy host, and then recovered from that experimentally induced illness. Enders’ own evidence makes it clear that even the first two steps (consistency in cases and isolation in pure culture) were not fully satisfied, and the final steps (disease reproduction and re-isolation) remained unperformed at the time his findings were publicly declared a discovery.

Disregarding the fact that the first two logical criteria were overlooked and unmet, the central question practically asks itself: Did Enders ever succeed in producing the disease in animals or humans using his cultured “measles agent,” as he himself said would be required for final proof? Did he ever provide the evidence he publicly acknowledged was necessary?

Let’s find out.

The Final Proof?

On May 28, 1955, The New York Times ran a headline that triumphantly declared: Enders Reproduces the Measles Virus. According to the report, Enders announced to 900 alumni at Harvard Medical School that he had “developed measles” in laboratory monkeys using tissue cultures derived from human cells. “Antibodies” were cited as confirmatory evidence of the findings:

“A modest scientist who shared the 1954 Nobel Prize in medicine for helping to isolate the poliomyelitis virus disclosed today that he had reproduced the measles virus. Dr. John F. Enders of Brookline, Mass., reported his new development to 900 alumni of the Harvard Medical School. He said, he had developed the measles in laboratory monkeys not treated with antibiotics. Cultures using human cells from tissue fragments were used in experiments, said Dr. Enders. Antibodies were employed to produce specific action against the suspected cause of the disease.”

However, Enders’ claimed “final proof” did not appear in print until 1957, in a study titled Behavior of Monkeys After Inoculation of Virus Derived from Patients with Measles and Propagated in Tissue Culture Together with Observations on Spontaneous Infections of These Animals by an Agent Exhibiting Similar Antigenic Properties. Although the full paper is not readily accessible, its findings can be reconstructed from contemporaneous accounts. In his 1956 Gordon Wilson Lecture, Enders summarized the results, claiming that inoculation of cynomolgus monkeys with a 23rd tissue-culture passage of the measles agent produced a disease that “in all essential respects closely resembles measles in man.” Notably, this outcome required combined intranasal and intravenous inoculation routes.

Enders presented detailed data from a single monkey as a representative example, while conceding that the full constellation of measles phenomena was not observed consistently. In several animals, the defining exanthem was absent or minimal, and in at least one case neither rash nor viremia occurred at all. Despite this variability, all inoculated monkeys were considered “infected” on the basis that each developed “antibodies:”

“Finally in cynomolgus monkeys a disease which in all essenitial respects closely resembles measles in man has followed inoculation of virus from the 23rd serial tissue culture passage of the virus. In Figure 10, the data obtained in one animal are presented as an example. The virus was introduced by both the intranasal and intravenous routes. Thereafter, no virus was recovered from the blood until the fifth day when it appeared and persisted until the 14th day. From the 11th to the 15th day a rash was observed on the face, trunk and extremities which closely resembled the exanthem seen in man. Within a day or two after the disappearance of the rash and within four days after the cessation of viremia, complement fixing antibody appeared in the blood which reached maximum titer within a few days. The concentration of this antibody declined rather rapidly during the ensuing weeks.

All these phenomena have not been seen in every monkey injected with the virus. In some the rash has been absent or minimal, although viremia has occurred. In one animal neither rash nor viremia were noted. In every case, however, antibody has developed.”

This admission is pivotal. Rather than demonstrating a fully reproducible disease caused by a purified, isolated agent, the experiment relied on passaged tissue-culture material administered via multiple artificial routes. Moreover, while Enders claimed that the disease “closely resembles” measles in humans, this does not equate to identical disease reproduction, and the full spectrum of clinical features was not observed. The experiment then broadened the definition of “infection” to include animals that lacked the hallmark clinical features of measles, such as the rash or viremia. “Antibody” production was therefore used as a surrogate marker for disease. As such, even this later work fails to meet Koch’s Postulates in substance: the disease was not recreated or consistently reproduced, the agent was never shown to be isolated and purified independently of the culture system, and causation was inferred rather than directly demonstrated.

Perhaps this explains why Enders concluded on a notably unconfident note. Rather than declaring that the measles agent had been definitively identified, he framed his conclusion conditionally, stating that if they were correct in believing they were in possession of the etiologic agent, then further work could proceed:

“In conclusion I would point out that if we are correct in thinking we are in possession of the etiologic agent of measles we have now the means of directly assaying the protective antibody in human globulin.”

This conditional phrasing is telling. It reflects an acknowledgment that the identification of the causative agent had not yet been firmly established, let alone conclusively proven. Rather than serving as a declaration of success, the statement functions as a provisional assumption—one that shifts attention away from demonstrating causation toward measuring “immune” responses. In effect, the existence of the agent is treated as a working hypothesis rather than a demonstrated fact, even as downstream applications are pursued.

Measles Recreated?

In the 1957 paper Measles Virus: A Summary of Experiments Concerned with Isolation, Properties, and Behavior, Enders was notably cautious in his description of the disease produced in monkeys, referring to it as a “mild disease comparable in most respects to measles in man.” That phrasing concedes three points simultaneously:

• The disease was mild, not measles as observed in humans
• It was comparable, not identical
• It matched measles only in “most respects,” not all

This description is problematic for any claim that the disease was truly reproduced. Koch’s Postulates require reproduction of the same disease, not a syndrome judged similar by qualitative comparison. By Enders’ own wording, that standard was not met.

More telling still are the conditions required to obtain even this limited result. Enders restricted his experiments to monkeys he classified as “susceptible,” inoculated them with unpurified, passaged tissue-culture material, and employed combined intranasal and intravenous routes of exposure. Even under these artificial and escalated conditions, the outcomes were inconsistent and variable.

Enders described the results as follows:

“The results of experiments in susceptible cynomolgus monkeys have indicated that two of the strains of virus tested after cultivation in vitro are capable of producing a mild disease comparable in most respects to measles in man. The criteria for determining the susceptibility of monkeys will be subsequently discussed. When animals were inoculated by both the intravenous and intranasal routes with virus of the first, second, and twenty-third tissue culture passages, several phenomena were noted. Viremia was established beginning on the fourth or fifth day and continued for two to five days. An exanthem appeared soon after the beginning of the viremic phase, extending over the thorax and abdomen and being especially prominent in the axillary and inguinal regions. Slight to moderate leukopenia developed and was maximal on the ninth day. Shortly after the period of exanthem—approximately two weeks after inoculation of the virus—specific complement-fixing antibodies appeared in the blood and soon attained maximal concentrations. These high levels were maintained for several weeks, after which they usually began to diminish. Antibodies, however, were detectable in significant amounts for at least eight months after inoculation of the virus. Not all of these phenomena were apparent in other animals inoculated at the same time with portions of the same virus preparation. In one monkey, no exanthem was observed, although viremia and antibody response were comparable to those of animals exhibiting a rash. Another responded only by the development of a slight leukopenia and the production of antibody. It is evident, then, that susceptible monkeys may vary considerably in their reaction to infection with this agent.”

In the 1960 paper Studies on an Attenuated Measles-Virus Vaccine — Development and Preparation of the Vaccine: Technics for Assay of Effects of Vaccination, Enders again stops short of claiming that measles itself was reproduced, describing instead “an infection resembling mild measles in man.” Even under highly artificial conditions—including combined intranasal and intravenous inoculation, and in some cases intracerebral and intracisternal injection into the brain, using tissue-culture material grown in human kidney cells—the defining clinical features of measles were inconsistently observed. Of the nine monkeys studied, viremia was detected in eight, yet the hallmark exanthem appeared in only four. This already limited and heterogeneous outcome is further constrained by the small sample size, which should have made broad conclusions about reproducibility or causation unwarranted. Despite this, all animals were classified as “infected” based on “antibody” responses, with serological patterns then compared to those seen in human measles convalescence:

“When cynomolgus monkeys, shown to be free of specific complement-fixing and neutralizing antibodies, were inoculated intranasally and intravenously or intracerebrally with early passages of the virus grown in human kidney-cell cultures an infection resembling mild measles in man and characterized by the following features developed: viremia; moderate leukopenia; exanthem; and emergence of neutralizing and complement-fixing antibodies. Although viremia was detected at some time between the third and the twelfth day after inoculation in all but 1 of 9 monkeys studied, rash was observed in only 4. Titers of complement-fixing antibody determined two and a half to four and a half weeks after inoculation ranged from 128* to 1024 in 8 of the animals. (The 9th succumbed earlier from unrelated causes.) Neutralizing-antibody titers recorded during the same period ranged from 90 to 1400. The concentrations of both antibodies were equivalent to or exceeded those found in the serums of recently convalescent patients. Antibodies declined gradually after attaining maximal titers, eventually to reach levels that tended to remain unchanged. In man the development, decline and persistence of measles antibodies appears to follow a similar pattern.”

The reasoning displayed by Enders substitutes “immunological” similarity for disease reproduction and allows “antibody” kinetics to serve simultaneously as evidence of “infection,” confirmation of the agent, and validation of the model. By Enders’ own numbers and wording, the disease was not fully or consistently reproduced, the exposure routes were non-physiological, and causation was inferred through pattern-matching rather than demonstrated through fulfillment of Koch’s Postulates.

“Infection” Without Illness

In 1962—nearly a decade after asserting that decisive proof would require recreating measles in animals and in humans—Enders acknowledged in Measles Virus: Historical Review, Isolation, and Behavior in Various Systems that the “pathogenicity of measles virus in laboratory animals has of late been little studied except in the case of monkeys” and deferred discussion of further results to a subsequent paper. This amounts to an admission that experimental reproduction of disease in animals was narrow, incomplete, and still pending.

Enders concession that the “pathogenicity” of measles “virus” in laboratory animals was “little studied,” was an admission that what evidence had been produced was neither clear nor decisive. This further indicates that claims of disease causation were not based on a robust experimental foundation across animal models, and that extrapolation to humans remained largely inferential rather than demonstrative.

When Enders did discuss animal “pathogenicity,” it depended on artificial adaptation, intracerebral inoculation, immature animals (suckling mice and hamsters), indirect serological identification, and the production of fatal encephalitic illness rather than measles itself. He ultimately concluded that even these findings were provisional, noting that final evaluation still awaited independent experimental confirmation:

“The pathogenicity of measles virus in laboratory animals has of late been little studied except in the case of monkeys. I shall outline results obtained in monkeys in a subsequent paper. Here we may mention the adaptation by Imagawa and Adams of the Edmonston strain of virus to suckling mice by intracerebral inoculation. On continuing passage a fatal illness was produced in the majority of animals inoculated. Identification of the passaged virus with the measles agent was indicated by neutralization tests carried out both in vitro and in vivo. These results are of interest in relation to those of Arakawa, who as early as 1948 described a fatal encephalitis in young mice inoculated with serum from early-phase measles sera. Arakawa’s mouse virus, which has subsequently been employed as vaccine before and after adaptation to chick embryo systems, has been reported by H. Taniguchi to exhibit cytopathogenicity in cell cultures comparable to that of the Edmonston strain. With the Edmonston strain, cross complement fixation reactions were also demonstrated. In suckling hamsters, Burnstein and his co-workers reported induction of a fatal infection by intracerebral passage of the virus. Taken as a whole these findings suggest that measles virus is pathogenic for immature mice and hamsters. Final evaluation must depend upon experimental confirmation by others.”

The subsequent paper discussing Ender’s work in relation to the “pathogenicity” of his tissue-culture agent in monkeys was titled Development of Attenuated Measles-Virus Vaccines. Rather than present evidence that he had been successful in his attempts to recreate measles in monkeys, in the paper Enders further illustrated the artificial and inferential nature of his measles research. Attempts to adapt early Edmonston “virus” strains to chick embryos initially failed, and successful replication required 28 passages in amnion cell cultures. Cytopathic effects in chick cells emerged only after several additional passages, demonstrating that observed “viral activity” was a product of extensive laboratory manipulation rather than natural behavior. When this “chick-adapted virus” was introduced into monkeys, no clinical signs of “infection”—rash, viremia, or “virus” recovery—were observed, yet all animals developed “antibodies.” Enders noted that these “antibodies” persisted for years in some cases, despite the absence of overt disease:

“Early in our work we attempted unsuccessfully to adapt five different strains of measles virus to the chick embryo. Included among them was the Edmonston virus. The materials tested were all derived from the first few passages after isolation in primate renal cell cultures. In spite of repeated trials, no evidence of the proliferation of these strains in the chick embryo was obtained. Recognition of spindle-cell transformation in amnion cell cultures suggested that a variant might have arisen; if so, we speculated that its capacity to multiply in chick systems might also be altered. Subsequent experiments did not clearly support this hypothesis. Nevertheless, this led us to inoculate chick embryos with fluid from the 28th amnion cell passage of the Edmonston strain. Viral multiplication was demonstrated without difficulty by tissue culture technique in the first and in all subsequent chick embryo passages. After six transfers, this chick-embryo-adapted virus was shown to be capable of continued multiplication in monolayer cultures of chick embryo cells. In this system cytopathic changes were not seen at first, but after the fourth passage spindle-cell transformation, small giant cells, rounding, and degeneration became regular features of the infectious process.

With the demonstration of these manifestations, suggesting rather profound changes in cytopathogenicity, it became of interest to determine the effect of the chick-adapted virus in monkeys. Experiment showed that a definite decrease in virulence for this host had been brought about during its sojourn in the chick system. Thus no rash or other signs of infection were observed. Failure of the chick-adapted virus to induce viremia detectable by the techniques employed was also in striking contrast to the constancy of the viremic phase after administration of the virulent kidney-cell progenitor. Moreover, in animals inoculated subcutaneously virus was not recovered from the pharynx, nor was it demonstrated in the spinal fluid or pharynx of others infected by the cerebral route, or in the cisterna magna of these animals. After this essentially inapparent infection, neutralizing and complement-fixing antibodies appeared consistently between the 14th and 21st days and reached levels comparable to those attained after inoculation of the virulent virus. In two animals reserved for prolonged study, these antibodies, after an initial decline, persisted in significant and unvarying concentrations over a period of 3 years.”

This reliance on serological markers as proof of “infection,” combined with non-physiological exposure routes and highly manipulated “viral” material, reflects a broader pattern in Enders’ work: disease was inconsistently reproduced, and causation was inferred indirectly rather than demonstrated. Even in later experiments, the defining clinical features of measles were not reliably reproduced, and claims of “pathogenicity” rested largely on serological surrogates rather than direct disease replication.

No Animal Model = No Recreation of Disease

Reiterating what he had revealed in his original measles paper in 1954, Enders acknowledged later that same year that measles had not yet been experimentally reproduced in animals and identified a fundamental obstacle in using “antibodies” to infer experimental measles “infection” in monkeys. In the Fourth R. E. Dryer Lecture Recent Observations on the Behavior in Tissue Culture of Certain Viruses Pathogenic for Man, he wrote:

“As yet, we have not reproduced the experimental disease by inoculation of these agents. Our attempt to do so in monkeys has been delayed by a rather unexpected but interesting finding. Complement fixation tests were done on a number of macacus rhesus and cynomolgus monkeys that had been held in captivity for considerable periods of time. To our surprise, nearly all these animals were found to possess antibody reacting often in high titer with the measles antigen. Two explanations immediately presented themselves. Either monkeys have a natural disease of their own, the agents of which possess antigens in common with that obtained from the human disease, or the animals contract measles upon exposure to human cases of measles before or after capture. The latter explanation appears to be probably correct. Thus, through the courtesy of the National Foundation for Infantile Paralysis and the Army Medical Service, we have recently been able to test 12 sera taken from cynomolgus monkeys 1 day after their capture in the jungle. No complement fixing antibody was found in any animal.”

This admission is critical: Enders openly conceded that experimental reproduction of measles had not been achieved and that “antibody” reactions—often present prior to inoculation—could not be assumed to reflect laboratory-induced “infection.” Serological responses depended on prior exposure and environmental factors rather than experimental intervention, rendering “antibodies” nonspecific and unreliable as confirmation of “infection.”

These limitations were independently reinforced by Russell J. Blattner in his 1957 review, Recent Developments in the Study of Measles Virus. He reported that many monkeys used in experiments already possessed appreciable titers of complement-fixing or neutralizing “antibodies” and noted cases of spontaneous “infection” in cynomolgus monkeys that closely resembled the response seen following inoculation. Importantly, he emphasized the possibility that a simian “virus,” antigenically indistinguishable from human measles strains, might exist naturally in monkey populations. Under such circumstances, serological responses could not discriminate between experimental inoculation, prior exposure, spontaneous “infection,” or cross-reactive agents. As Blattner observed, these findings were of “considerable significance,” highlighting the lack of controlled experimental proof that the cultured “virus” introduced in the laboratory was responsible for the observed effects. “Antibody” responses were treated circularly—assumed to indicate “infection” because “infection” was presumed—which left fundamental causal questions unresolved.

Even historically, researchers noted that “infection” of monkeys with clinical material from measles patients could produce signs resembling—but not identical to—human measles. However, “infection” with measles “virus” isolated and propagated in cultured cells did not consistently induce these clinical signs (Enders et al., 1960; Yamanouchi et al., 1970; van Binnendijk et al., 1994; Nii et al., 1964; Sakaguchi et al., 1986). This distinction highlights a persistent difficulty: laboratory-propagated material failed to reliably reproduce the disease it was presumed to represent.

These limitations continued to affect later experimental work. Enders’ attempts to induce measles in monkeys using passaged tissue-culture “virus” required artificial and invasive routes of exposure (intranasal, intravenous, intracerebral and intracisternal), and even then, symptoms were inconsistent and serological markers nonspecific. By all accounts, he did not provide the “final proof” that would satisfy Koch’s Postulates: the disease was not consistently reproduced, the agent was not shown to be isolated independently of tissue culture, and causation remained inferred rather than demonstrated.

The historical record reflects this unresolved problem, where multiple researchers failed to transmit measles to monkeys. As early as 1919, virologist Andrew Sellards concluded that there was “no exact” or “convincing proof of the susceptibility of monkeys” to measles. In fact, he felt that one of the cardinal problems remaining to be solved for measles was the experimental recreation of the disease in animals, despite decades of unsuccessful attempts. This helps explain why, even today, a fully suitable animal model for measles has not been established.

Modern reviews confirm this continuity. A 1996 National Institutes of Health guide confessed that “measles is a difficult virus to study because there are no satisfactory animal models.” A 2009 review of measles animal models noted that “animal models are highly important to understand the pathologic mechanisms of viral diseases” and that “the lack of a suitable animal model has greatly hindered the research into the pathogenesis of measles.” This concern was echoed in 2014 by Richard Plemper, PhD, who stated that “there is no good animal model for testing measles virus infection.” Most recently, a 2022 study confirmed the ongoing limitation, observing that “the lack of animal models that are not only permissive to MeV, but also reproduce infection of human diseases may explain the scarcity of protective efficacy studies for these diseases.” As there are no satisfactory animal models, then the very experiment Enders said was required for final proof could not—and did not—occur. At best, measles research proceeded without the decisive causal test Enders himself deemed necessary.

Taken together, Enders’ historical experiments, Blattner’s observations, and contemporary assessments illustrate a clear continuity: despite decades of work, a reliable animal model that faithfully reproduces human measles remains lacking, and serological markers alone are insufficient to demonstrate causation. These persistent limitations prevented Enders and later researchers from providing definitive proof that the measles “virus” caused the disease, leaving causal claims inferred rather than demonstrated. This gap continues to influence modern measles research.

Game Over

Crucially, the evidence Enders presented falls short of the definitive proof he himself acknowledged was necessary, undermining claims that he had successfully isolated the measles “virus” or conclusively established its “pathogenicity.” Enders was explicit about what would constitute definitive proof. In his 1954 measles paper, he stated that it was essential to reproduce the disease experimentally in both animals and man. Yet his animal experiments repeatedly failed to produce a consistent or faithful reproduction of human measles, and he never even attempted inoculation in man. This omission is especially damaging given that, in his Gordon Wilson Lecture, Observations on Certain Viruses Causing Exanthematous Diseases in Man, Enders conceded that “absolute proof can be obtained only by inoculation of man.” By his own stated criteria, then, Enders never achieved absolute proof that he had isolated or identified the measles “virus” at all.

Enders’ “game” was one of substitution and misdirection. He treated laboratory artifacts—cytopathic effects in cultured cells—as proxies for genuine “viral” activity, played nonspecific “antibody” responses as surrogate markers of “infection,” and moved experimentally inoculated animals around a board of artificial exposure routes, counting inconsistent or incomplete disease patterns as victories. Evidence that should have called his claims into question—CPE in uninoculated cultures, nonspecific serology, and variable or absent disease in monkeys—was ignored or reinterpreted to maintain the narrative of success. Yet, by presenting these inferential and provisional results as definitive proof—ignoring the very criteria he himself had outlined as essential to establish said proof—Enders effectively reset the rules of the game, and his unscientific playbook became a foundational strategy for modern virology.