https://sriranjith.github.io/ A curated, carefully sourced collection of paradoxes, laws, and effects that shaped how we think — explained for every age, from a first story to the original papers. en https://sriranjith.github.io/paradox/amaras-law/ https://sriranjith.github.io/paradox/amaras-law/ Tue, 01 Sep 2026 12:00:00 GMT We overestimate the effect of technology in the short run and underestimate it in the long run. The internet would transform everything — in two years. Ten years later it had done less than promised. Thirty years later it had done far more than anyone predicted. The pattern repeats for every major technology. Economics Computer Science AI Philosophy https://sriranjith.github.io/paradox/anchoring-effect/ https://sriranjith.github.io/paradox/anchoring-effect/ Tue, 01 Sep 2026 12:00:00 GMT The first number you hear pulls all your subsequent estimates toward it — even when it is arbitrary. Shown a wheel landing on 65, people estimate Africa makes up 45% of the UN. Shown one landing on 10, they estimate 25%. The number was meaningless. The 20-point gap was not. Psychology Economics https://sriranjith.github.io/paradox/baader-meinhof-phenomenon/ https://sriranjith.github.io/paradox/baader-meinhof-phenomenon/ Tue, 01 Sep 2026 12:00:00 GMT Learn a new word, hear a new concept, buy a new car — and suddenly it seems to appear everywhere. The world hasn't changed. Your perception has. Two cognitive biases cooperate: selective attention makes you notice the thing once familiar, and confirmation bias logs each occurrence as evidence it really is everywhere. Psychology Philosophy https://sriranjith.github.io/paradox/brandolinis-law/ https://sriranjith.github.io/paradox/brandolinis-law/ Tue, 01 Sep 2026 12:00:00 GMT The energy needed to refute misinformation is an order of magnitude larger than the energy needed to produce it. A false claim takes seconds. Refuting it requires evidence, context, and defence against follow-up objections. The asymmetry structurally favours misinformation. Philosophy Society Computer Science https://sriranjith.github.io/paradox/chestertons-fence/ https://sriranjith.github.io/paradox/chestertons-fence/ Tue, 01 Sep 2026 12:00:00 GMT Do not remove a fence until you understand why it was built. If you cannot explain why it exists, you do not understand it well enough to know what removing it will break. Seeming pointlessness is evidence of ignorance, not evidence of pointlessness. Philosophy Society Economics https://sriranjith.github.io/paradox/cunninghams-law/ https://sriranjith.github.io/paradox/cunninghams-law/ Tue, 01 Sep 2026 12:00:00 GMT The best way to get the right answer on the internet is not to ask a question — it is to post the wrong answer. People who would not bother helping someone who asked politely will spend considerable effort correcting someone who is confidently wrong. The correction instinct is stronger than the helping instinct. Psychology Society Computer Science https://sriranjith.github.io/paradox/dunbars-number/ https://sriranjith.github.io/paradox/dunbars-number/ Tue, 01 Sep 2026 12:00:00 GMT The human brain can maintain stable social relationships with approximately 150 people — the same count found in hunter-gatherer bands, Roman army companies, and Christmas card lists. Beyond that, groups need formal rules to stay coherent. The limit is cognitive, built into the primate neocortex. Psychology Biology Society https://sriranjith.github.io/paradox/flynn-effect/ https://sriranjith.github.io/paradox/flynn-effect/ Tue, 01 Sep 2026 12:00:00 GMT IQ scores have been rising in virtually every country measured, by roughly 3 points per decade. The gains are too large and fast to be genetic. Something in the environment is making IQ test performance steadily improve. We don't fully know what, or what it means. Psychology Biology Society https://sriranjith.github.io/paradox/metcalfes-law/ https://sriranjith.github.io/paradox/metcalfes-law/ Tue, 01 Sep 2026 12:00:00 GMT The value of a network is proportional to the square of its users. Double the network, quadruple the value. This explains why network effects are so powerful — why Facebook moated itself as it grew, why telephone adoption accelerated, and why networked markets tend toward monopoly. Computer Science Economics AI https://sriranjith.github.io/paradox/pareto-principle/ https://sriranjith.github.io/paradox/pareto-principle/ Tue, 01 Sep 2026 12:00:00 GMT Roughly 80% of consequences come from 20% of causes. Eighty percent of a country's wealth is held by 20% of its population. Eighty percent of software bugs come from 20% of the code. The ratio varies, but the principle — that a small minority of causes drives the vast majority of effects — recurs everywhere. Economics Mathematics Statistics https://sriranjith.github.io/paradox/placebo-effect/ https://sriranjith.github.io/paradox/placebo-effect/ Tue, 01 Sep 2026 12:00:00 GMT A sugar pill, a saline injection, sham surgery — if a patient believes they are receiving treatment, they often improve. Not just subjectively: measurable physiological changes occur. The belief that you are being treated can produce real healing. Psychology Biology https://sriranjith.github.io/paradox/ringelmann-effect/ https://sriranjith.github.io/paradox/ringelmann-effect/ Tue, 01 Sep 2026 12:00:00 GMT As group size increases, each individual exerts less effort. Two people pulling a rope don't pull twice as hard as one; eight people don't pull eight times as hard. The effect is not just coordination: individuals genuinely work less hard when others are present. Psychology Society Economics https://sriranjith.github.io/paradox/semmelweis-reflex/ https://sriranjith.github.io/paradox/semmelweis-reflex/ Tue, 01 Sep 2026 12:00:00 GMT New knowledge contradicting established beliefs is rejected reflexively, even when overwhelming. Semmelweis showed handwashing reduced childbed fever deaths from 10% to under 2%. The medical establishment dismissed him, and his discovery went unused for a generation while thousands died of preventable infections. Psychology Society Philosophy https://sriranjith.github.io/paradox/sunk-cost-fallacy/ https://sriranjith.github.io/paradox/sunk-cost-fallacy/ Tue, 01 Sep 2026 12:00:00 GMT We keep investing in failing projects because of what we have already spent — even though past investment cannot be recovered and should be irrelevant. The rational rule is to decide on expected future costs and benefits. The psychological reality is we cannot stop throwing good money after bad. Psychology Economics https://sriranjith.github.io/paradox/zipfs-law/ https://sriranjith.github.io/paradox/zipfs-law/ Tue, 01 Sep 2026 12:00:00 GMT In almost any large text, the most common word appears twice as often as the second most common, three times as often as the third. The same distribution governs city populations, website traffic, income, and earthquakes — with no obvious reason why they should all obey the same rule. Mathematics Statistics Probability https://sriranjith.github.io/paradox/benfords-law/ https://sriranjith.github.io/paradox/benfords-law/ Mon, 31 Aug 2026 12:00:00 GMT In almost any large collection of real-world numbers — populations, prices, river lengths, stock prices — the digit 1 appears as the leading digit about 30% of the time. The digit 9 appears only 5% of the time. Expected each digit equally at 11%? Almost every dataset proves you wrong, and the law catches fraudsters. Mathematics Statistics https://sriranjith.github.io/paradox/brooks-law/ https://sriranjith.github.io/paradox/brooks-law/ Mon, 31 Aug 2026 12:00:00 GMT Adding manpower to a late software project makes it later. New developers need ramp-up time — during which existing developers must train them. More people also multiplies communication channels quadratically. The project falls further behind exactly when adding people feels like the fix. Computer Science Economics Psychology https://sriranjith.github.io/paradox/bystander-effect/ https://sriranjith.github.io/paradox/bystander-effect/ Mon, 31 Aug 2026 12:00:00 GMT The more people witness an emergency, the less likely any one of them is to help. Not because they do not care, but because each person assumes someone else will act — and the responsibility, spread across the crowd, falls through the cracks entirely. The safest place to need help is an empty street. Psychology Society https://sriranjith.github.io/paradox/cobra-effect/ https://sriranjith.github.io/paradox/cobra-effect/ Mon, 31 Aug 2026 12:00:00 GMT A solution designed to reduce a problem ends up increasing it, because the incentive rewards the very thing it was meant to eliminate. Named from the British cobra bounty in colonial India: citizens farmed snakes for the reward, and when the scheme was cancelled, released them — leaving more cobras than before. Economics Society Psychology https://sriranjith.github.io/paradox/conways-law/ https://sriranjith.github.io/paradox/conways-law/ Mon, 31 Aug 2026 12:00:00 GMT Any organisation that designs a system produces a design whose structure mirrors the organisation's communication structure. Teams build systems that reflect how they talk to each other — not by plan, but because the natural division follows the organisational boundary. Computer Science Society Psychology https://sriranjith.github.io/paradox/dunning-kruger-effect/ https://sriranjith.github.io/paradox/dunning-kruger-effect/ Mon, 31 Aug 2026 12:00:00 GMT People with limited skill in a domain overestimate their ability; the highly competent underestimate theirs. The cause is recursive: the skills needed to perform well are the same ones needed to recognise good performance. Low performers lack the tools to see their own incompetence. Psychology Philosophy https://sriranjith.github.io/paradox/halo-effect/ https://sriranjith.github.io/paradox/halo-effect/ Mon, 31 Aug 2026 12:00:00 GMT A single striking trait — beauty, confidence, a firm handshake — floods our judgment of everything else about a person. We rate the likeable candidate as more competent, the confident speaker as more accurate, even when none of those things follow. The halo is invisible to the person wearing it. Psychology https://sriranjith.github.io/paradox/hanlons-razor/ https://sriranjith.github.io/paradox/hanlons-razor/ Mon, 31 Aug 2026 12:00:00 GMT Never attribute to malice that which is adequately explained by incompetence. When something goes wrong, the malicious explanation requires extra assumptions — that someone chose to cause harm — that are usually unnecessary. Most harm comes from error, not evil, and the right diagnosis produces better remedies. Philosophy Psychology Society https://sriranjith.github.io/paradox/hawthorne-effect/ https://sriranjith.github.io/paradox/hawthorne-effect/ Mon, 31 Aug 2026 12:00:00 GMT People change their behaviour when they know they are being observed. Workers in a 1920s factory became more productive with every experimental change — including when changes were reversed — simply because someone was paying attention. The measurement itself changes what is being measured. Psychology Statistics Biology https://sriranjith.github.io/paradox/hofstadters-law/ https://sriranjith.github.io/paradox/hofstadters-law/ Mon, 31 Aug 2026 12:00:00 GMT It always takes longer than you expect, even when you take Hofstadter's Law into account. This self-referential formulation captures why estimation bias is not correctable by simply knowing about it: you underestimate the correction too. The effect is consistent and found across virtually every domain of complex work. Psychology Computer Science Mathematics https://sriranjith.github.io/paradox/matthew-effect/ https://sriranjith.github.io/paradox/matthew-effect/ Mon, 31 Aug 2026 12:00:00 GMT Small early advantages compound into large inequalities, even when later performance is equal. The researcher cited first gets cited more; the child who reads earliest reads more and learns faster. Named from Matthew 25: "To everyone who has, more will be given; from him who has not, even what he has will be taken." Society Economics Psychology https://sriranjith.github.io/paradox/moores-law/ https://sriranjith.github.io/paradox/moores-law/ Mon, 31 Aug 2026 12:00:00 GMT Since the integrated circuit, transistor counts on a microchip have roughly doubled every two years while costs halved. Gordon Moore noticed this in 1965 and predicted it would continue. It did — for fifty years. The pace is now slowing, but the law reshaped every industry on Earth while it held. Computer Science AI Economics https://sriranjith.github.io/paradox/murphys-law/ https://sriranjith.github.io/paradox/murphys-law/ Mon, 31 Aug 2026 12:00:00 GMT Anything that can go wrong, will go wrong — and at the worst possible moment. Stated as a joke at an Air Force test site in 1949, it spread into engineering culture as a serious design principle: if a component can be installed wrong, someone will install it wrong, so design it to allow only the correct way. Psychology Computer Science Philosophy https://sriranjith.github.io/paradox/occams-razor/ https://sriranjith.github.io/paradox/occams-razor/ Mon, 31 Aug 2026 12:00:00 GMT Among competing explanations that fit the evidence equally, prefer the one with the fewest unnecessary assumptions. The razor does not guarantee truth — the simplest explanation is often wrong — but it is the correct starting prior when evidence does not yet distinguish between competing theories. Philosophy Mathematics Computer Science https://sriranjith.github.io/paradox/parkinsons-law/ https://sriranjith.github.io/paradox/parkinsons-law/ Mon, 31 Aug 2026 12:00:00 GMT Work expands to fill the time available for its completion. A task that could take an hour, given a week, will take a week — not because the work grows, but because anxiety, iteration, and perfectionism fill the container. Stated as satire about bureaucracy in 1955, it has since been supported by research. Psychology Economics Society https://sriranjith.github.io/paradox/pygmalion-effect/ https://sriranjith.github.io/paradox/pygmalion-effect/ Mon, 31 Aug 2026 12:00:00 GMT When someone in authority expects you to succeed, you tend to — not because they predicted correctly, but because their expectation changes how they treat you, which changes what you become. The teacher who believes a student is bright creates a brighter student. The prophecy fulfils itself. Psychology Society https://sriranjith.github.io/paradox/streisand-effect/ https://sriranjith.github.io/paradox/streisand-effect/ Mon, 31 Aug 2026 12:00:00 GMT Attempting to suppress information reliably causes it to spread far more widely. The suppression attempt itself becomes the news, attracting far more attention than the original content. Named after Barbra Streisand's 2003 lawsuit to remove aerial photos — which caused millions to seek them out. Psychology Society Computer Science https://sriranjith.github.io/paradox/butterfly-effect/ https://sriranjith.github.io/paradox/butterfly-effect/ Mon, 31 Aug 2026 12:00:00 GMT A butterfly flapping its wings in Brazil might set off a tornado in Texas — not by force, but because tiny differences in initial conditions compound into completely different outcomes. The future is fundamentally unpredictable past a short horizon, even in a perfectly deterministic universe. Physics Mathematics Philosophy https://sriranjith.github.io/paradox/peter-principle/ https://sriranjith.github.io/paradox/peter-principle/ Mon, 31 Aug 2026 12:00:00 GMT In a hierarchy, every employee tends to rise to their level of incompetence. A person good at their job is promoted; if good at that, promoted again — until reaching a position they cannot handle, where they stay. The result: senior positions are disproportionately occupied by people unequipped to hold them. Psychology Economics Society https://sriranjith.github.io/paradox/achilles-and-the-tortoise/ https://sriranjith.github.io/paradox/achilles-and-the-tortoise/ Sat, 29 Aug 2026 12:00:00 GMT Achilles gives a tortoise a head start. By the time he reaches where it was, it has crawled a little further — and so on, forever. Zeno concluded he can never overtake it. He does, and the infinitely many stages take a perfectly ordinary eleven seconds. Philosophy Mathematics https://sriranjith.github.io/paradox/arrows-impossibility-theorem/ https://sriranjith.github.io/paradox/arrows-impossibility-theorem/ Sat, 29 Aug 2026 12:00:00 GMT Write down four things any fair voting system should obviously do. Arrow proved in 1951 that with three or more options, no method of combining ranked preferences can do all four — except one that always follows a single person. Economics Society Mathematics https://sriranjith.github.io/paradox/berksons-paradox/ https://sriranjith.github.io/paradox/berksons-paradox/ Sat, 29 Aug 2026 12:00:00 GMT Two unrelated things can look strongly related once you only look at a selected slice. Among hospital patients, among the people you date, among everyone you hired — the selection itself invents the correlation. Statistics Probability Psychology https://sriranjith.github.io/paradox/berrys-paradox/ https://sriranjith.github.io/paradox/berrys-paradox/ Sat, 29 Aug 2026 12:00:00 GMT "The smallest positive integer not definable in under sixty letters." That phrase has fifty-seven letters, so it defines the very number it says cannot be defined. The way out is that "definable" cannot mean what it seems to. Logic Mathematics Computer Science https://sriranjith.github.io/paradox/bertrands-box-paradox/ https://sriranjith.github.io/paradox/bertrands-box-paradox/ Sat, 29 Aug 2026 12:00:00 GMT Three boxes, two drawers each: two gold coins, two silver, one of each. You open a drawer and find gold. The chance the other drawer in that box is also gold is 2/3, not 1/2 — and this is exactly the Monty Hall problem, posed a century earlier. Probability Mathematics https://sriranjith.github.io/paradox/braess-paradox/ https://sriranjith.github.io/paradox/braess-paradox/ Sat, 29 Aug 2026 12:00:00 GMT Open a fast new road between two points and every driver's journey gets longer. Close a busy street and the traffic speeds up. Dietrich Braess proved in 1968 that this is not bad luck — it is what selfish route choice does. Economics Computer Science Society https://sriranjith.github.io/paradox/buridans-ass/ https://sriranjith.github.io/paradox/buridans-ass/ Sat, 29 Aug 2026 12:00:00 GMT A donkey stands exactly midway between two identical bales of hay. Having no reason to prefer either, it starves. The story is named after a man who never told it, and it is really a question about what a reason is for. Philosophy Psychology https://sriranjith.github.io/paradox/cantors-diagonal-argument/ https://sriranjith.github.io/paradox/cantors-diagonal-argument/ Sat, 29 Aug 2026 12:00:00 GMT Hand over any list claiming to contain every infinite sequence of digits. Reading down the diagonal and changing each entry produces a sequence that is not on it. No list can be complete, so some infinities are bigger than others. Mathematics Logic https://sriranjith.github.io/paradox/gabriels-horn/ https://sriranjith.github.io/paradox/gabriels-horn/ Sat, 29 Aug 2026 12:00:00 GMT Rotate the curve y = 1/x around the x-axis from x = 1 to infinity. The solid of revolution has a finite volume — you can fill it with paint — but an infinite surface area — you cannot paint its inside. The same amount of paint does both and neither. Mathematics https://sriranjith.github.io/paradox/galileos-paradox/ https://sriranjith.github.io/paradox/galileos-paradox/ Sat, 29 Aug 2026 12:00:00 GMT Most whole numbers are not perfect squares, so the squares must be rarer. Yet every number has exactly one square and every square has exactly one root, so there are precisely as many of one as of the other. Both claims are true. Mathematics Philosophy https://sriranjith.github.io/paradox/godels-incompleteness/ https://sriranjith.github.io/paradox/godels-incompleteness/ Sat, 29 Aug 2026 12:00:00 GMT Any consistent set of rules rich enough to describe arithmetic contains true statements it cannot prove — and cannot prove its own consistency. A precise theorem about formal systems, and the most misquoted result in mathematics. Logic Mathematics Computer Science Philosophy https://sriranjith.github.io/paradox/goodharts-law/ https://sriranjith.github.io/paradox/goodharts-law/ Sat, 29 Aug 2026 12:00:00 GMT Pick a number that stands in for something you care about, then reward people for moving that number. They will move it. The thing you actually cared about is now free to drift away, and your number will not tell you that it has. Economics AI Society https://sriranjith.github.io/paradox/hilberts-hotel/ https://sriranjith.github.io/paradox/hilberts-hotel/ Sat, 29 Aug 2026 12:00:00 GMT A hotel with infinitely many rooms, every one occupied, can still take in a new guest — and a coachload, and infinitely many coachloads. Being full stops meaning what it means everywhere else. Mathematics Philosophy https://sriranjith.github.io/paradox/levinthals-paradox/ https://sriranjith.github.io/paradox/levinthals-paradox/ Sat, 29 Aug 2026 12:00:00 GMT A protein chain has more possible shapes than there are atoms in the galaxy. If it tried them one at a time it could not find the right one before the universe ended. It finds it in millionths of a second, every time. Biology Physics Computer Science https://sriranjith.github.io/paradox/maxwells-demon/ https://sriranjith.github.io/paradox/maxwells-demon/ Sat, 29 Aug 2026 12:00:00 GMT A tiny doorkeeper sorts fast molecules from slow ones and makes a hot room and a cold room out of a lukewarm one, for free. It took 115 years to find the bill: the demon must eventually erase its own memory, and erasing costs heat. Physics Computer Science https://sriranjith.github.io/paradox/menos-paradox/ https://sriranjith.github.io/paradox/menos-paradox/ Sat, 29 Aug 2026 12:00:00 GMT If you already know what you are looking for, you do not need to search for it. If you do not know what you are looking for, you will not recognise it when you find it. So inquiry is either unnecessary or impossible. Yet we learn, and we search, and we recognise discoveries when we make them. Philosophy