Consider the phrase “it was meant to be.”（“这是命中注定的” in Chinese）When someone says this after a failure or a stroke of luck, they are making a claim that sounds meaningful. It has the texture of an explanation. It offers comfort. And yet, if you press on it for even a moment, it dissolves into pure circularity: what happened was always going to happen, because it happened. This is a TAUTOLOGY.
This is not a small logical slip. It is a complete abdication of explanatory power. The philosopher Karl Popper gave us the precise tool to understand why: a theory that can explain every possible outcome explains nothing at all. If “destiny” can account for both success and failure, both action and inaction, both belief and disbelief — then it carries zero information about the world. It is, in the language of information theory, a message with no information.
This essay begins with that intuition — the irritation at unfalsifiable thinking — and takes it somewhere more unsettling. Because the same failure mode that makes “destiny” intellectually worthless does not only appear in folk philosophy or astrology. It appears in science itself. Not in science’s ideals, but in its practice: in the institutions, the consensus-makers, the gatekeepers who decide what counts as a signal and what gets dismissed as noise.
The argument here is not anti-science. It is pro-science in the most demanding sense: science’s greatest virtue is its willingness to be wrong. The tragedy is that this virtue is systematically undermined by two structural failures embedded in how science actually operates. Understanding those failures is not optional — it may be a matter of survival.
In 1934, Karl Popper published the German edition of Logik der Forschung; the expanded English edition, The Logic of Scientific Discovery, appeared in 1959. And with it, a deceptively simple criterion that would become the foundation of modern philosophy of science: a theory is scientific if and only if it can, in principle, be proven wrong.
The intuition behind falsifiability is elegant. Science does not accumulate truth by confirming what it already believes. It advances by surviving attempts at refutation. The more a theory sticks its neck out — the more specific and risky its predictions — the more we learn from its survival. A theory that predicts nothing specific can never be defeated, but for exactly that reason, it tells us nothing useful.
This maps cleanly onto information theory. The information content of a claim is inversely proportional to its probability: I(A) = log(1/P(A)). A statement that is always true — P(A) = 1 — carries zero information. This is not a metaphor. It is a mathematical statement about the epistemic value of predictions. “Destiny”, with P(A) approaching 1 for every conceivable outcome, is not merely vague — it is formally empty.
Einstein understood this better than almost anyone. When he published the General Theory of Relativity in 1915, he did not merely offer a new framework for understanding gravity. He made a specific, dangerous prediction: that light from distant stars would bend as it passed near the sun. This prediction was testable. It was falsifiable. And crucially, if the 1919 solar eclipse expedition had returned with null results, Einstein’s theory would have been dead.
That willingness to stake everything on a single observation is the gold standard of science. It is what separates a genuine intellectual commitment from a rhetorical performance. A theory that cannot be killed cannot be trusted.
And yet, here is the problem. This is the ideal. What follows is the practice.
If falsifiability is the engine of scientific progress, what happens when that engine stalls? The history of science is not only a story of brilliant theories and brave predictions. It is also a story of harm: real, physical harm inflicted on real people, while the scientific establishment looked away, or worse, looked directly at the evidence and denied it.
Two distinct failure modes are responsible. They are not the same failure, and they deserve to be named separately.
Science requires sufficient evidence before updating its consensus. This is, in principle, a feature. Anecdote should not overturn established knowledge. Single cases are unreliable. Statistical significance exists for good reasons.
But here is the structural problem: harm does not wait for statistical significance. While science accumulates the sample size it needs to reach a conclusion, the people generating that sample are suffering. In the worst cases, they are dying. The epistemological machinery of science is correct in its requirements, but indifferent to the cost of meeting them. Those early victims are not just unfortunate — they are, in a cold and precise sense, the tuition that science charges for updating its beliefs.
This is the first failure: not malice, but method. A lag that is built into the structure of how knowledge advances.
The second failure is more troubling, because it is not a limitation of method but a corruption of spirit. When anomalous signals appear — symptoms that do not fit the model, harms that the theory cannot explain — the scientific establishment does not always respond with curiosity. It responds with dismissal.
Real suffering gets relabeled as “psychosomatic.” Consistent reports from hundreds of individuals get filed under “anecdotal.” The patients who cannot be explained become patients who are not believed. This is not the scientific method. This is the scientific establishment using the prestige of the method as a shield against the very falsification the method demands.
In Popper’s terms, this is immunization: the act of protecting a theory from refutation by explaining away every piece of contrary evidence rather than taking it seriously. It is exactly what “destiny” does. And it is exactly what scientific institutions have done, repeatedly, throughout history.
The following cases are not selected to be sensational. They are selected because each one illustrates, with varying degrees of finality, the two failure modes described above.
In the 1910s and 1920s, radium was not merely considered safe — it was marketed as a health product. Radium-laced water was sold in elegant bottles. Radium was added to toothpaste, face cream, and chocolate. At the United States Radium Corporation in New Jersey, young women were employed to paint watch dials with radium-based luminescent paint. They were instructed to point their brushes with their lips.
The women began to fall ill. Teeth fell out. Jawbones disintegrated. Bones across their bodies developed spontaneous fractures. Some of them glowed faintly in the dark — their very skeletons saturated with radioactive material. They were dying, and they knew what was killing them.
The scientific and corporate establishment’s response was Failure 2 in its purest form. Company-appointed doctors attributed symptoms to syphilis — a diagnosis that conveniently implied the women’s own moral failing. Independent researchers who raised concerns found their funding threatened. The science was not absent. It was actively recruited to suppress the truth.
A major 1928 settlement in New Jersey came early, but the broader legal and institutional reckoning unfolded over the following decade, with the 1938 Donohue case often treated as a decisive milestone. By then, many were dead. Their suffering was not a mystery waiting for science to catch up: their bodies were generating the data in real time. The lag was not methodological. It was moral.
In 1949, the Nobel Prize in Physiology or Medicine was awarded to Egas Moniz for the development of the prefrontal leucotomy — a procedure in which a metal instrument was inserted through the eye socket and used to sever connections in the frontal lobe. The procedure was later popularized in the United States by Walter Freeman, who performed it with an ice pick, often without anesthesia, in what he called “transorbital lobotomy.”
Approximately 40,000 to 50,000 lobotomies were performed in the United States alone. The results included severe personality changes, loss of initiative, emotional blunting, incontinence, and in many cases, a state that contemporary observers described as turning patients into “zombies.” Families reported that the person who returned from surgery was not the person who had entered.
Physicians who raised doubts were marginalized. The Nobel Prize had conferred a kind of institutional immunity on the procedure. The signal was there — in the patients sitting vacantly in institutions, in the families who had lost someone — but the authority of the prize created a gravitational field that bent professional judgment around the truth rather than toward it.
The lobotomy is now recognized as one of the most catastrophic chapters in the history of medicine. It is also a case study in how scientific prestige can function as an instrument of suppression rather than a vehicle for truth.
Not every case of scientific dismissal ends in death. Some end in a quieter indignity: the systematic disbelief of people reporting genuine experience.
Synesthesia — the neurological phenomenon in which stimulation of one sense automatically triggers another, so that sounds have colors, letters have personalities, or numbers occupy specific spatial positions — was reported by individuals for centuries. For much of the twentieth century, it was classified as a symptom of mental illness, drug use, or simple confabulation. Neurologists who took it seriously were regarded as credulous.
The modern scientific consensus, established through neuroimaging in the 1990s and 2000s, is that synesthesia is a genuine, heritable neurological variant with measurable neural correlates. It is real. It was always real. The people reporting it were not confused or lying. The scientific establishment simply lacked the instrumentation to detect what was happening, and in its absence of tools, defaulted to dismissal.
Synesthesia illustrates the gentler face of Failure 2: not the violent suppression of the Radium Girls, but the quiet pathologization of experience that does not fit the existing model. The cost here was not physical death but epistemic erasure — the invalidation of a person’s inner life by an institution that had no tools to see it.
The cases above are settled. The science has spoken, the harm has been confirmed, and the institutions that suppressed the truth have been judged by history. But the lesson of those cases is precisely that at the time they were unfolding, they did not look settled. They looked like fringe concerns, anecdotal complaints, the kind of thing that serious scientists did not bother with.
There are two ongoing disputes worth placing on the record here — not as established facts, but as live questions that the framework above should make us take seriously.
The first is the question of OLED displays and eye health. Hundreds of users across forums and medical communities report persistent eye strain, headaches, and visual disturbances that correlate with extended exposure to OLED screens. The mechanism proposed by some researchers involves PWM (pulse-width modulation) flicker at frequencies that conventional testing does not measure under normal conditions. The scientific consensus currently holds that approved displays are safe. Whether this consensus reflects the truth or reflects the limits of current measurement instrumentation is a question that, given the history above, deserves more than dismissal.
The second is low-frequency noise and its effects on human health. In China and elsewhere, residents in proximity to certain industrial equipment, ventilation systems, and infrastructure report debilitating symptoms: persistent headaches, sleep disruption, anxiety, and a pervasive sense of pressure or vibration that they cannot localize. When they report these symptoms to authorities, the response is often that standard noise measurement equipment detects nothing. “Non-actionable” becomes the official classification. But the measurement equipment operates within a defined range, and the human body does not. The absence of a reading is not the same as the absence of a signal. In China, the regulatory framework has not kept pace with the reports, leaving thousands of people with no recourse and no acknowledgment.
These cases may be resolved in either direction. They are placed here not as verdicts but as illustrations of a pattern: the pattern of a measurement apparatus that defines the boundaries of what is real, and a regulatory and scientific establishment that mistakes those boundaries for the boundaries of reality itself.
If you had lived in New Jersey in 1922, the science would have told you that radium was safe. If you had been a psychiatric patient in 1947, the science would have told you that lobotomy was a valid and effective treatment. If you had reported seeing colors when you heard music in 1970, the science would have told you that something was wrong with you.
In each case, the science was wrong. And in each case, the wrongness was not a matter of insufficient data waiting to be collected. The data was there. The signals were being generated by human bodies in real time. What was missing was not evidence. What was missing was the willingness to take it seriously.
This is not a counsel of despair, and it is not an invitation to distrust science. It is an argument for holding science to its own highest standard. Popper was right: the willingness to be wrong is what makes science powerful. The tragedy is that scientific institutions are composed of humans, with human tendencies toward authority-preservation and consensus-maintenance, and those tendencies systematically suppress exactly the willingness that makes science valuable.
The question is not whether science is better than the alternatives. It is. The question is: given what we know about how scientific consensus can fail — given the Radium Girls, the lobotomies, the synesthetes dismissed as delusional — how confident should you be that the current consensus on any given question is correct?
You will not know which current belief is wrong in the way the people in those cases did not know. That is the nature of the problem. The best you can do is to take your own body and your own experience seriously, to resist the impulse to defer to authority when authority asks you to disbelieve what you are directly perceiving, and to remember that the history of science is, in part, a history of people who were right before the institutions were willing to admit it.
The dead do not get to update their priors. The rest of us still can.