Why Myths Multiply When Randomness Is Left Unexplained

Bad explanations breed fastest where uncertainty is real, consequences are personal, and nobody has been taught the grammar of chance.

This is not a small problem of old aunties, temple gossip, WhatsApp forwards, horoscope columns, miracle cures, political rumor, or the gentleman at the tea stall who can explain monsoon failure, cholesterol, exam results, share-market crashes, and geopolitics with one eyebrow and no evidence. Those are symptoms. The deeper problem is representational. Human beings are forced to live inside systems too large, too noisy, too interconnected, and too delayed in their consequences to be understood by instinct alone. When the mind meets a foggy system, it does not politely wait for Bayesian inference [Bayesian inference, a method of updating beliefs when new evidence arrives]. It makes a story. Often a bad one. Sometimes a dangerous one. Occasionally a story wearing sandalwood paste, laboratory language, nationalism, family prestige, or management consultancy perfume.

A myth is not merely a false belief. It is a low-cost compression of anxiety. A superstition is not merely stupidity. It is a control ritual invented under uncertainty. A just-so story is not merely a lazy explanation. It is a narrative that pretends the current arrangement of things was inevitable, meaningful, or morally deserved. The child failed because Saturn was badly positioned. The business collapsed because someone cast an evil eye. The fever broke because the amulet worked. The neighbor’s son got a job because he was “meant for success.” The poor are poor because they lack discipline. The rich are rich because they are brilliant. The patient died because “God wanted it.” The city floods because people have become sinful, or because one party is evil, or because one engineer was corrupt, when the actual system may include drainage geometry, wetland destruction, silt, unplanned construction, solid waste, colonial-era infrastructure, monsoon intensity, groundwater behavior, institutional neglect, procurement incentives, and nobody being accountable for the whole elephant.

Randomness is the first insult to the storytelling mind. It says that not every event is a message. Not every coincidence is a clue. Not every streak is destiny. A person may smoke for fifty years and live to ninety, while another person may never smoke and still develop lung cancer. This does not mean smoking is harmless. It means individual outcomes are noisy samples from distributions. A distribution is the pattern formed by many possible outcomes, not the fate of one person. The human mind, unfortunately, is a village lawyer arguing from a sample size of three. It remembers the uncle who drank, smoked, abused everyone, and lived forever. It forgets the silent cemetery of people who did not.

This distinction matters in Calcutta because the city is a factory of small data. Everyone knows someone. Everyone has a case. Everyone has a cousin, tutor, doctor, astrologer, broker, priest, political contact, or coaching-center survivor who proves something. Dense social life is wonderful for affection and disastrous for inference. It turns anecdotes into evidence with the speed of hot oil catching mustard seeds. “My neighbor took this medicine and recovered” becomes pharmacology. “My friend invested here and doubled money” becomes economics. “That family did this ritual and their daughter got married” becomes cosmology. “That boy never studied and cracked the exam” becomes education policy. A city rich in stories can become poor in causality.

Randomness does not mean the world is meaningless. It means causes do not always announce themselves neatly at the scale of individual experience. Many causes are probabilistic. They change odds rather than guarantee outcomes. A vaccine does not create an invincible shield. It shifts risk. A public health intervention does not abolish disease. It changes transmission probability. A good school does not guarantee wisdom. It changes exposure, discipline, peer effects, language access, and opportunity. A fair hiring process does not guarantee perfect selection. It reduces structural bias. The ordinary person does not need to become a statistician, but without this basic sense of probability, life becomes a theatre of false certainty.

Chaos is the second insult. In ordinary speech, chaos means mess. In mathematics, chaos means something more interesting and more treacherous: a deterministic system whose future can become practically unpredictable because tiny differences in starting conditions grow rapidly over time. The weather is the polite textbook example. Calcutta traffic is the sweaty one. A bus stopping half a minute longer near Gariahat, a handcart angled badly near Burrabazar, one broken signal near Esplanade, one impatient biker sliding into a gap that should have remained empty, and suddenly the city has composed a symphony in honking and despair. Nobody planned the jam. Nobody fully owns it. Yet everyone contributes.

Chaotic systems are not magic. They are lawful systems that outrun simple prediction. This is a crucial distinction. If a system is chaotic, shouting “everything happens for a reason” is not explanation; it is wallpaper. The reason may exist, but the chain may be too sensitive, too long, and too interdependent to reconstruct cleanly. A family argument, a hospital delay, a market crash, a rumor, a stampede, a dengue outbreak, a sudden price spike, an online mob, or a political panic can all behave this way. The system has rules. The outcome still surprises everyone, including the people who helped cause it.

Complexity is the third insult, and perhaps the most important. A complex system is not merely complicated. A jet engine is complicated. Remove a part and a trained engineer can often say what will fail. A city is complex. A hospital is complex. A family is complex. A healthcare network is complex. A rumor ecosystem is complex. In a complex system, many agents interact, adapt, learn, deceive, cooperate, overload, improvise, and route around constraints. The result is emergent behavior, meaning the whole does things that cannot be understood by staring only at the parts. A colony of ants is not just many ants. A market is not just many buyers. A hospital is not just doctors, nurses, beds, medicines, software, and billing counters. It is a living arrangement of formal processes and informal survival tactics.

This is where myths become dangerous rather than quaint. Myths simplify complex systems by assigning blame or virtue to the nearest visible object. The patient worsened because the doctor was careless. Sometimes true. Often incomplete. The doctor may be trapped inside a brittle Electronic Health Record [EHR, the clinical system used to document patient care], incomplete medication history, delayed lab reporting, missing allergies, overcrowded wards, weak handoff notes, exhausted staff, contradictory family accounts, antibiotic overuse, poor follow-up, reimbursement pressure, and a culture where nobody admits uncertainty because uncertainty is mistaken for incompetence. The visible person becomes the lightning rod for an invisible architecture.

In healthcare Information Technology [IT, the systems used to store, move, secure, and analyze healthcare information], this failure appears constantly. A lab result arrives late and people call it a data quality problem. Sometimes it is. But often the deeper failure is representational. The system captured the result but not the workflow state. It stored the value but not the clinical context. It preserved the timestamp but not what the timestamp means. Was it specimen collection time, accession time, result time, verification time, interface transmission time, or time of display in the receiving system? These are not clerical details. They are the bones of reality. When representation fails, downstream users call the data “bad,” as if the data were a naughty schoolboy. The data may be faithfully carrying a malformed model of the world.

The same mistake happens in ordinary life. A person sees an outcome and mislabels the representation failure as a moral fact. A student’s low score becomes laziness, when the hidden variables may include language barriers, sleep, nutrition, household stress, bad teaching, exam design, anxiety, class signaling, or simply variance. A shop’s decline becomes incompetence, when the system includes platform economics, delivery subsidies, real estate pressure, changing consumer habits, procurement scale, algorithmic visibility, and customers trained to value speed over neighborhood trust. A marriage collapses and the family explains it through horoscope mismatch, when the actual system includes gendered labor, money, emotional illiteracy, migration, status anxiety, and two people who were never taught how to disagree without turning into minor deities of vengeance.

The non-obvious architectural insight is this: myths persist not because people lack intelligence, but because myths are often cheaper than maintaining a high-fidelity model of causality. Accurate models are expensive. They require data, patience, humility, revision, and tolerance for “we do not know yet.” Myths require none of these. They arrive fully dressed, emotionally satisfying, and usually with a villain. In a society under pressure, myth is an energy-saving device. It reduces cognitive load. It also ruins decisions.

The difference between data transport and semantic meaning is essential here. Transport is the movement of a signal from one place to another. Meaning is what that signal validly represents inside a context. Health Level Seven version 2 [HL7 v2, a widely used healthcare messaging standard for transmitting events such as admissions, discharges, orders, and results] can move a message from a laboratory system to an EHR. That does not mean the receiving system understands the clinical meaning with full fidelity. Fast Healthcare Interoperability Resources [FHIR, a modern healthcare data standard built around modular resources and web-based exchange] can expose structured data through an Application Programming Interface [API, a controlled way for software systems to communicate]. That does not mean the data is semantically clean. A rumor moving through WhatsApp has excellent transport. Its meaning may be garbage wearing shoes.

This distinction is not only technical. It is civic. Calcutta has superb rumor transport. Tea stalls, family groups, para networks, coaching circles, political clubs, apartment associations, local pharmacies, temples, mosques, clinics, and mobile phones form a dense mesh of transmission. But semantic validation is weak. Who knows? How do they know? What is the denominator? What is the counterexample? What incentives shape the claim? What would change our mind? These questions are the civic equivalent of schema validation. Without them, the city becomes an interface engine for nonsense.

A just-so story is especially seductive because it explains backward. It begins with the current state and invents a path that makes the outcome feel natural. The successful person was always destined. The failed person had warning signs. The powerful institution was always legitimate. The oppressed group somehow deserved its position. The old custom must contain deep wisdom because it is old. The new technology must be progress because it is new. This is how societies launder accident into destiny. It is also how they protect bad architecture.

Complexity teaches a harsher lesson: outcomes are often produced by interactions, not essences. A person is not simply “disciplined” or “undisciplined.” Behavior emerges from incentives, constraints, habits, environments, relationships, fatigue, fear, opportunity, and feedback. A hospital is not simply “efficient” or “inefficient.” It behaves according to bed capacity, staffing ratios, software design, referral patterns, procurement rules, clinical governance, billing models, regulatory reporting, and the thousand unofficial workarounds by which humans prevent formal systems from collapsing. A city is not simply “cultured” or “decaying.” It is a sedimentary system, with empire, capital, migration, politics, climate, language, infrastructure, education, and memory packed into it like geological strata, except the rocks also argue on Facebook.

The practical implication is that design and governance must begin by asking what kind of uncertainty is being handled. If the problem is random variation, collect enough observations and reason in probabilities. If the problem is chaotic sensitivity, monitor early signals and avoid overconfident long-range prediction. If the problem is complex adaptation, design feedback loops, incentives, accountability boundaries, and graceful failure modes. In healthcare IT, that means preserving provenance, modeling workflow states explicitly, distinguishing event time from record time, avoiding premature normalization when meaning is still contested, and refusing to treat interoperability as mere message movement. In ordinary life, it means asking whether a claim is based on a distribution, a mechanism, a tradition, a fear, a sales pitch, or one spectacular uncle.

There is, however, a realistic constraint that prevents a clean solution. People do not live as detached analysts. They live tired, indebted, overworked, status-anxious, family-bound, politically manipulated, medically under-informed, and algorithmically provoked. A person waiting in a crowded clinic with a sick parent does not have the luxury of reading a monograph on uncertainty. A student facing exams does not calmly model variance. A shopkeeper watching customers vanish into quick-commerce platforms does not run a complexity simulation before despairing. Good explanations must therefore be usable under stress. They must be short enough to remember and strong enough to resist nonsense.

One useful rule is this: a single story is not a system. When someone explains a large outcome through one cause, be suspicious. One community is not poor because of one trait. One city is not broken because of one party. One disease surge is not caused by one festival. One child’s success is not caused by one coaching center. One stock did not rise because the universe rewarded optimism. Complex systems usually have braided causes. Pulling one thread and declaring victory is how bad policy, bad medicine, and bad family advice are born.

Another rule: coincidence is not communication. The mind loves pattern, because pattern recognition helped our ancestors survive. But the same gift also makes us see agency in noise. A phone call arrives after a dream. A black cat crosses the road before bad news. A ritual precedes recovery. A politician visits a shrine before winning. A patient starts three treatments and improves. Which one worked? Maybe one. Maybe none. Maybe the body healed. Maybe the disease fluctuates naturally. Maybe the outcome was already likely. Without comparison, timing seduces us into false causality.

A third rule: always ask about the missing cases. Who is absent from the story? Who failed silently? Who was never counted? In healthcare data, missingness is not emptiness; it often has meaning. A missing lab value may mean the test was not ordered, could not be afforded, was performed elsewhere, failed processing, was not interfaced, or was clinically irrelevant. In social life, missing cases are even more treacherous. We hear from survivors, winners, loud relatives, visible families, successful migrants, and confident frauds. We do not hear from the people who followed the same advice and sank quietly.

A fourth rule: incentives are often stronger than declarations. If a system rewards engagement, it will produce outrage. If a school rewards exam rank, it will narrow learning. If a hospital rewards volume, it will optimize throughput. If a political machine rewards identity panic, it will manufacture enemies. If a software project rewards go-live dates over semantic correctness, it will ship brittle interfaces and call the wreckage modernization. Complexity does not abolish responsibility. It locates responsibility more accurately.

None of this requires contempt for tradition. Some inherited practices encode observation, hygiene, ecology, social coordination, or psychological comfort. The mistake is not having rituals. Humans need rituals. The mistake is confusing ritual with mechanism, comfort with evidence, antiquity with truth, and identity with causality. A grandmother’s remedy may soothe. It may even work for a mild condition. But if it delays antibiotics for sepsis, insulin for diabetic ketoacidosis, or psychiatric care for severe depression, then culture has crossed from meaning into harm. The world is not obliged to honor our metaphors.

For the ordinary Bengali living in Calcutta today, randomness, chaos, and complexity are not elite abstractions. They are survival tools. They help decide whether to believe a medical claim, how to interpret a child’s marks, why the city floods, why traffic collapses, why markets punish small shops, why political rumors spread, why smart people make foolish predictions, and why personal virtue does not guarantee worldly success. They do not make life simple. They prevent false simplicity from becoming tyranny.

The point is not to replace old superstition with new technocratic arrogance. Models also fail. Data can lie by omission. Algorithms can scale historical injustice. Experts can be captured by institutions, incentives, career fear, or fashionable nonsense. Scientific language can become just another priestly costume if it refuses scrutiny. The cure for myth is not blind faith in numbers. It is disciplined doubt, better representation, open correction, and the moral patience to say, “This is what we know, this is what we do not know, this is what would change our view.”

A society that cannot reason about chance becomes cruel to the unlucky. A society that cannot reason about chaos becomes addicted to prediction gurus. A society that cannot reason about complexity becomes easy prey for demagogues, miracle sellers, fake healers, bad software, bad policy, and family tyrants with confident voices. Calcutta, with its books, gossip, humidity, brilliance, decay, tenderness, and daily improvisations, deserves better than explanation by cosmic shrug.

The mature mind does not stop telling stories. It tells better ones. It keeps the story porous enough for evidence, humble enough for uncertainty, and precise enough to guide action. It knows that a map is not the territory, a message is not meaning, a correlation is not a cause, and a loud explanation is not necessarily a true one. That is not the death of wonder. It is wonder with better plumbing.