Insight Editions · No. 31
The Upright Thinkers
The Human Journey from Living in Trees to Understanding the Cosmos
Leonard Mlodinow
Leonard Mlodinow recounts the development of scientific thought, from early human cognition and ancient philosophy to modern physics, chemistry, biology, and the structure of the atom.
A study guide to The Upright Thinkers by Leonard Mlodinow. Shared free for personal reading; not for sale and not endorsed by the authors or publishers. Please read and buy the original.
Chapter 1. Our Drive to Know
The book opens in Buchenwald. Mlodinow's father, who weighed 85 pounds at liberation, once traded a crust of bread for the answer to a math puzzle a fellow prisoner had posed. To the son, that trade is the purest evidence that the search for knowledge is among the most human of desires. The title makes the same point: our ancestors stood upright, which freed hands and gaze, "but above all, we are thinkers."
A second anecdote sets the tone. As a writer on Star Trek: The Next Generation he once pitched a scientifically careful idea and was told by a producer to shut up, in stronger words, and tell a story. The lesson stuck: science is best understood as a human story, with motives, rivalries and luck.
| Measure | Then | Later | |
|---|---|---|---|
| Top travel speed | Camel caravan, a few mph (cities, about 4000 B.C.) | Chariot about 20 mph; steam 100 mph by 1900; Space Shuttle 17,000 mph | Time to go from 10 to 100 mph |
| About 2 million years | 50 years from 100 to 1,000 mph | Years to reach 75% of US homes | Landline phone: 81 |
From this he draws the structure of the book. Part I spans millions of years (the brain, the "why" question, writing, mathematics, law, philosophy). Part II spans hundreds (Galileo, Newton, Lavoisier, Darwin). Part III spans decades (the quantum). The economist Kenneth Boulding observed that the world of his old age was as different from the world of his birth as that world was from Julius Caesar's.
Mlodinow also insists on the unity of knowledge. Science did not spring from a few isolated geniuses; its roots run through Greek philosophy, religion, art, alchemy, medieval universities and even postal systems. Reuters began, he notes, with carrier pigeons. The history of physics, chemistry and biology is therefore also a history of trade routes, institutions and the people who kept records.
- It is hard to conceive of what does not yet exist. Even Isaac Asimov's Foundation novels of the 1950s imagined interstellar empires in which men commuted to offices and women stayed home.
- It is hard to accept change once it is conceived. New ideas meet resistance from the people they displace, and from habits of thought. As Einstein put it, science is "as subjective and psychologically conditioned as any other branch of human endeavor."
The acceleration has a cost, which links the two themes. In a world that changes slowly, a person's education lasts a lifetime; in one that changes every decade, it expires. Mlodinow's own career illustrates the point: trained as a theoretical physicist, he moved into television, video games and books, each time relearning the trade. The Reuters news agency began in 1850 by using carrier pigeons to bridge a gap in the telegraph line between Aachen and Brussels; within a few years the gap closed and the pigeons were obsolete. The skill that lasts is the capacity to keep learning.
Mlodinow's producer on Star Trek had a point. Ideas are easier to grasp and remember when they come with people, stakes and surprises. Each chapter here follows the same pattern: a puzzle, the person who wrestled with it, what blocked them, and what finally worked. That structure is also a good model for explaining technical change inside an organisation.
Bread, speed and the warrant
Mlodinow opens with a story that is not a metaphor. His father, starving in Buchenwald, met an emaciated inmate educated in mathematics who would not give away a puzzle's answer. The man held the secret "as if it were a hunk of gold." Amid stench and death the father became obsessed with knowing. Eventually the deal was struck: a crust of bread for the solution. After American forces liberated him, the father weighed eighty-five pounds. The son's career in theoretical physics sits on that inheritance: wanting to know can override wanting to eat.
The chapter then widens to historical speed. From Lucy (about 3.2 million years ago) to camel caravans at a few miles per hour took almost the whole of human evolution. From steam at about 100 mph around 1900 to spacecraft at tens of thousands of miles per hour took decades. Household adoption compresses the same curve: landline phones needed about 81 years to reach three-quarters of US homes; cell phones about 28; smartphones about 13. Acceleration is not a slogan. It is the shape of the record.
- It is hard to conceive of what does not yet exist. Galileo's peers lacked the category "mathematical law of free fall"; they were not merely bad at arithmetic.
- It is hard to accept change once someone has conceived it. Phlogiston chemists, anti-Darwinian clergy and Einstein resisting quantum probability all illustrate the second theme.
He is not claiming every century is happier than the last. He is claiming that the toolkit for asking why compounds. A smartphone is not wiser than Thales; it embeds Thales's heirs in silicon and glass. When the book later reaches quantum theory, the same pattern returns: people could not imagine probability as a fundamental law, and once someone could, many of the best minds refused it.
| Era | Top travel speed (book's figures) | Comment | |
|---|---|---|---|
| About 4000 B.C. | a few mph (camel caravan) | Millions of years from Lucy | Chariot age |
| about 20 mph | Still animal power | Steam, by 1900 | about 100 mph |
| Industrial jump | Space Shuttle | about 17,000 mph | Decades, not millennia |
The personal thread returns at the end of Part I and in the epilogue. Laws of physics comfort because, unlike the decrees that seized a tailor's shop in Częstochowa in 1939, they cannot be rewritten by a conquering army overnight. That is an emotional claim with an analytical bite. Science is not only predictive power. It is a public, revisable kind of authority.
Before a debate, ask: is this a conception problem (no one has the category yet) or an acceptance problem (the category exists and threatens someone)? Conception needs prototypes and analogies. Acceptance needs coalitions, demos and time.
Teams sometimes "pay bread" (ship date, political capital) for a number they could have instrumented for free. The upright move is to price the question: what is the cheapest observation that would settle it? If the answer costs a crust and the ignorance costs a quarter, buy the crust.
| Signal | Drive to know | Mere distraction | |
|---|---|---|---|
| Cost willingly paid | Comfort, status, even food | Only spare time | Stopping rule |
| A decisive test | Novelty fades | Social form | Shared puzzle, open method |
| Gossip, hot take | Legacy | A tool others inherit | A mood |
Read Chapter 1 as the book's warrant. If you do not believe humans have a drive to know that sometimes beats hunger, later chapters about Galileo or Darwin will look like hobbyism. If you do, they look like the drive meeting institutions. Keep a scorecard: for each major figure, what could not be conceived before them, and who refused after them.
Working the opening harder
Return to the Buchenwald bargain with the chapter's quieter claims. The inmate would not say the answer at first; he made the father discover that the need itself had a price. Mlodinow does not solve the puzzle for the reader. He solves the motivation. Science biographies that skip this motive become lists of clever people. This book is an argument that the motive is species-typical and historically accelerating.
The speed tables are not trivia. Going from roughly 10 mph thinking to 100 mph vehicles took on the order of two million years; going from 100 to 1,000 took about fifty. That asymmetry is why Part III can happen in decades after Parts I and II took millennia. When your org debates "moving fast," ask which layer you are on: biological curiosity, cultural ratchet, civilizational tools, or instrumented theory. Each has a different clock.
| Episode | Conception hard | Acceptance hard | |
|---|---|---|---|
| Galileo's free fall | Speed as number | Church + Aristotle professors | Oxygen |
| Gas as substance of combustion | Phlogiston loyalty | Natural selection | Selection without a selector |
| Religious and social status | Quantum probability | Chance as fundamental | Einstein and realists |
Product teams often use a biological clock (sprint anxiety) for a civilizational problem (data model redesign). Name the clock before you name the deadline. Upright thinkers match the tool layer to the tempo.
The Częstochowa thread is not only memoir. It is a control case for authority: human law flipped in days; physical law did not. When a bank or platform rewrites policy overnight, users feel the first kind of authority. When a checksum still fails the same way after a regime change, they feel the second. Build more of the second into systems that claim to be fair.
Bread for a puzzle. In Buchenwald an inmate traded his crust for a maths answer. Drive to know can outweigh hunger. When a team stalls, ask: what would we trade comfort to learn?
Pick one stubborn open question at work (a failed metric, an unexplained outage). Write the smallest experiment that would change your mind in either direction. Run it this week, or schedule it. Do not add a second experiment until the first reports.
Curiosity without a stopping rule becomes scope creep. The same hunger that bought a puzzle can burn weeks chasing trivia.
Drive, bread and two clocks
Father’s bread-for-puzzle sets scarcity-grade curiosity. Invention time and acceptance time are different clocks. Science history pairs conceivers with cultures that metabolise slowly. Acceleration compresses invention and can freeze acceptance behind credentials and dashboards.
| Clock | Measures | Failure | |
|---|---|---|---|
| Invention | how fast a new idea appears | clever dead ends | Acceptance |
Acceptance lag as the plot
Read every later chapter as a pair: who conceived the change, who accepted it, what it cost refusers and refugees. Tools, teaching and incentives each speed or freeze the stack.
Curiosity that only performs at dinner is not the chapter’s point. Survival-grade questions change what you do next.
Chapter 2. Curiosity
A lizard confronted with a box cannot reason about what is inside; it can only act on instinct. The story of the genus Homo is the story of a creature that could. Our line runs back to a half-pound placental mammal (the book names Protungulatum donnae) that survived the six-mile-wide asteroid that killed about 75% of species 66 million years ago. It took 60 million years to reach the apes, a few million more for the human body, and only about 10,000 years for the cultural changes that built the modern world.
Julian Jaynes wrote that at some point human evolution "turned at a right angle," from changes in the body to changes in the mind. Mlodinow's version: physical evolution takes millions of years, mental capacity a few hundred thousand, and cultural change only thousands. The pace of the book follows that curve.
Brains evolved first to control movement: more than half our neurons serve motor control and the senses. The upright posture freed the hands and changed the gaze, and with it the problems the brain could work on.
In June 2017 two papers in Nature (Hublin et al.; Richter et al.) reported Homo sapiens fossils from Jebel Irhoud, Morocco, dated by thermoluminescence of heated flints to 315,000 ± 34,000 years ago. They pushed the origin of our species back by about 100,000 years and suggest a pan-African origin rather than a single East African cradle. Lucy, meanwhile, marked the 50th anniversary of her discovery (24 November 1974) in 2024.
Imagine a street on which each house holds one generation of your ancestors, and drive backward along it. You would have to drive almost 4,000 miles to reach Lucy, a 3-foot-7, 65-pound hairy "woman" who would look more like a chimp than a relative. About halfway there you would pass Homo habilis, some 100,000 generations removed from Lucy.
Brains are expensive. The brain is about 2% of body weight but burns about 20% of our calories, while muscle uses about a tenth as much energy per unit mass. Size within a species does not predict brilliance: Byron's brain reportedly weighed about 5 pounds, Anatole France's about 2 and Einstein's 2.7. Daniel Lyons, who died in 1907 at 41 with normal body weight and normal intelligence, was found at autopsy to have a brain of only 680 grams, about a pound and a half. What counts is organisation, and above all the kinds of question the brain is inclined to ask.
- Chimp strength. The book repeats the claim that a chimpanzee can pull over 1,200 pounds. A 2017 PNAS study (O'Neill et al.) found chimpanzee muscle produces about 1.35 times the force and power of human muscle of the same size: strong, not superhuman.
- The bottleneck. The book says modern humans shrank to a few hundred individuals about 140,000 years ago. Population-genetics estimates vary widely; the best-known recent claim (Hu et al., Science, 2023) places a severe bottleneck of about 1,280 breeding individuals far earlier, 930,000 to 813,000 years ago, and that result is itself disputed by other geneticists.
Each step in the lineage bought new mental territory. Homo habilis, named by the Leakey team in the early 1960s, made simple stone flakes; bonobos trained for years cannot learn to make comparable tools, and brain imaging suggests toolmaking recruits a left-hemisphere network close to the one used for language. Homo erectus hunted, probably controlled fire and travelled out of Africa; its males and females were closer in size than earlier species, a hint of pair bonding and shared provisioning. The common thread is that each species faced harder problems and the brain grew to meet them, at a steep energy cost.
The heart of the chapter is the "why" question. In 1918, on Tenerife, Wolfgang Köhler watched chimpanzees stack boxes to reach bananas: real problem-solving. But in a modern study, children given a trick L-shaped block that would not stand up turned it over to see why; chimps did not. In the 1920s a researcher logged 40 "why" questions in four days from one four-year-old. Apes taught sign language answer questions; they never ask them. Even infants have expectations about physics: at the University of Illinois infant cognition lab, babies of about six months look longer when a larger cylinder rolling down a ramp pushes a toy bug a shorter distance than a smaller one did. Expectation, surprise and the urge to explain are built in.
Mlodinow treats the urge to explain as an evolved trait, not a luxury. The species that asks "why does it fall?" can predict where it will fall. The same logic applies to organisations: teams that ask why a process works (not only how) are the ones that notice when it stops working.
Curiosity as hardware
Chapter 2 argues that science rests on a brain shaped by survival pressures, not by a taste for textbooks. Australopithecines like Lucy had brains about a third the size of ours; Homo habilis about twice Lucy's and half of ours. Stone tools appear about 2 million years ago. Fire, hunting in groups and, much later, language layered new uses onto old circuits. The point is not that "we evolved to do physics." The point is that the same machinery that tracks predators and food can be pointed at invisible causes.
Mlodinow's lizard story is the everyday version. In Southern California his son Nicolai hunted small lizards that freeze, then flee. A large inverted box plus cardboard turned the hunt into a capture. Humans do the freeze-then-flee sequence with status threats and bad dashboards. Science interrupts the freeze: name the fear, invent a test, update. Children ask "why" before they can spell; adults must keep asking after the social cost rises.
| Stage | Relative brain size (book) | Rough date | |
|---|---|---|---|
| Lucy ( A. afarensis ) | 1× | about 3.2 million years ago | Homo habilis |
| about 2× Lucy | about 2 million years ago | Modern Homo sapiens | about 4× Lucy |
The chapter quietly corrects a romantic picture of the lone observer. Nicolai's hunts only work with a gadget plus a habit. That foreshadows Galileo's ramp, Boyle's pump and the LHC: curiosity becomes science when tools force nature to answer in numbers. Over millions of years, minds triumphed over obstacles that once demanded only strength and agility of bodies.
Animals explore. What humans add is shared models that outlive the explorer. A lone genius with no students, no tablets and no instruments leaves a dead end. Chapter 3 names the accumulation a cultural ratchet; Chapter 2 supplies the motor.
Predator logic: freeze, then flee, conserve energy. Lab logic: approach the anomaly with a controlled probe. Hiring systems that punish false starts recreate predator logic inside research orgs.
On-call engineers often freeze on the first alarming graph. Train the opposite: a 90-second write of feared cause, alternative cause and discriminating check. Then run the check.
| Instinctive | Scientific | ||
|---|---|---|---|
| Trigger | Threat, novelty | Anomaly vs model | First move |
| Freeze or flee | Instrument and compare | Memory | Individual |
| Shared lab notebook | Success metric | Survival today | Prediction tomorrow |
For readers in 2026 the fossil timeline has shifted (see the Update on Jebel Irhoud), but the cognitive claim does not depend on a 200,000-year species age versus a 300,000-year one. Whenever Homo sapiens fully assembled, the why-asking habit was already being selected and then amplified by culture.
Tools, freeze and the reuse of old circuits
Chapter 2's evolutionary sketch can be misread as genetic destiny. Mlodinow's actual claim is weaker and more useful: circuits built for survival scanning can be redirected to invisible causes if culture supplies tools and norms. The lizard hunt is a parable of instrumentation. Without the box, curiosity ends in a blur of motion. With the box, curiosity becomes data.
Stone tools at about 2 million years are the first great "box." Fire, language and later writing are bigger boxes. Each box lets a why-question survive the asker's death. That is why brain size ratios matter less than tool plus teaching. A larger brain without a ratchet still resets.
The Lucy / habilis / sapiens ratios are not IQ scores. They are Mlodinow's way to say: the hardware upgraded, yet the software of science arrived late and socially. Do not hire for "brain size." Hire for probe design.
- Feared cause (one sentence).
- Alternative cause (one sentence).
- Discriminating check (one action, time-boxed).
- Only then escalate or patch. Teams that skip to patch practice predator logic with a pager.
When a benchmark drops, teams freeze on the leaderboard narrative. Apply Chapter 2: is the drop measurement drift, data mix shift or model regression? Instrument before you retrain. Retraining is fleeing.
Chapter 3 without Chapter 2 is empty corridors. Keep both.
Curiosity’s invoice and freeze
Lucy, habilis, erectus: posture, tools, range. Brains cost calories. Infants run occlusion/support/continuity labs on a schedule. Freeze-then-flee is the threat default; science names the threat and cheap-tests. Organisations punish failed probes and reverse the infant schedule.
Built-in labs and organisational freeze
Metabolic-brain percentages differ by model; the qualitative invoice stands. Celebrate questions only with time and access budgets.
Red metric: write feared cause, alternate cause, cheapest separating observation. Run the observation before the all-hands narrative.
Narratives feel like action; probes are action.
Lucy, costly brains and infant physics
Lucy and the bipedal line free the hands; Homo habilis and erectus push tools and range. Brains are metabolically expensive: the book stresses the calorie invoice for a larger cortex. Curiosity is not free. Infants fail and pass naive physics tests on a schedule that looks like a built-in lab course: occlusion, support, continuity. The "why" question is the species' signature once language can carry it.
If your org celebrates questions but does not budget calories (time, compute, access), you are praising Lucy's posture while starving the brain.
Freeze, then flee. Lizards freeze when threatened, then bolt. Humans freeze on bad news too. Science is the trained habit of not freezing: name the threat, then move toward a test.
Next time a dashboard turns red, write three lines before you act: (1) what you fear happened, (2) what else could explain it, (3) the cheapest observation that separates (1) from (2). Then act on (3).
Curiosity that never commits becomes analysis paralysis. Hominins who only froze starved. A hypothesis you will not test is decoration.
Chapter 3. Culture
Once self-aware, humans noticed their own mortality, and asked who was in charge of the cosmos. Self-awareness had a dark side. Humans are perhaps the only animals that know they will die, and the oldest burials, grave goods and cave art suggest that early people asked existential questions: who made the world, who controls the weather, what happens after death. Those questions, Mlodinow argues, are the ancestors of science as much as of religion. Both try to find order and agency behind events.
Mlodinow's thesis about the Neolithic revolution (the term was coined by V. Gordon Childe in the 1920s) is that it was mental and spiritual first, agricultural second.
The old story was that climate or population pressure forced foragers to farm. The evidence fits poorly. Skeletal studies from the 1980s show pre-farmers were well fed, while early farmers had worse teeth, spines and anemia, and died younger. The San (Bushmen) foraged more efficiently than pre-war European farmers, working perhaps two to four hours a day for a "varied and abundant" diet. So why settle?
Near Urfa in southeast Turkey, on a hill whose name means "potbelly hill," hunter-gatherers about 11,500 years ago (7,000 years before the Great Pyramid) raised rings of carved T-shaped pillars up to 18 feet tall, some stones weighing 16 tons, without metal tools, the wheel or draft animals. A 1960s survey dismissed the site as a Byzantine cemetery; a farmer's plough struck a stone in 1994, and Klaus Schmidt excavated until his death in 2014. The carvings show lions, snakes, boars and scorpions. To Ian Hodder, it showed that "sociocultural changes come first, agriculture comes later."
The book follows Schmidt in calling it a sanctuary with no habitation. Excavations led since 2014 by Lee Clare (German Archaeological Institute) found domestic buildings, hearths, grinding stones, a large rock-cut cistern and rainwater channels; DAI now describes a settlement with a strong ritual component. Clare's 2026 palaeoclimate paper (Heritage) places densest occupation about 8700 to 8200 calBC, with rectangular/trapezoidal EPPNB dwellings overlapping the monumental phase, and links later disruption to rapid climate change, slope slips and possible quakes. In 2026 Necmi Karul's Taş Tepeler team exposed more rectangular second-phase buildings north of the famous enclosures; residential use is the working impression pending floor-sediment analysis. Whether the earliest PPNA phase was permanently occupied remains open. The core point survives: monumental, communal building came before farming.
The foragers' leisure is documented. In the 1960s the anthropologist Richard Lee lived with the !Kung San of the Dobe area in the Kalahari and found adults spent roughly two to three days a week getting food, with a diet that was varied and reliable even in drought years. Farming, by contrast, meant longer hours, a narrower diet and dependence on a few crops. Settling down was not an obvious improvement in living standards, which is why the motives must have been something other than hunger.
| Measure | Foragers | Early farmers | |
|---|---|---|---|
| Hours of food work | About 2 to 4 a day | Longer, seasonal peaks | Diet |
| Varied (dozens of plant and animal species) | Few staple grains | Health from skeletons | Taller, better teeth |
| Shorter, more anemia, dental decay, spinal wear | Population density | About 1 person per square mile | Villages of one to two dozen dwellings, later thousands |
The phrase holds for Göbekli Tepe's builders, who hunted gazelle and aurochs and gathered wild cereals. But the line between foraging and farming in the region was blurry: wild wheat was harvested intensively nearby, and domesticated einkorn has been genetically traced to the Karacadağ hills only about 30 km away. Monuments and the first steps toward farming may have grown together rather than one strictly preceding the other.
Çatalhöyük (about 7500 B.C., central Turkey) housed up to 8,000 people but, Mlodinow argues, was not a city, because it lacked division of labour. Its people buried their dead under the floors and painted bull-baiting murals, which he reads as a new attitude of dominion over animals that preceded domestication of sheep, goats, cattle and pigs over about 2,000 years.
- The book says Homo floresiensis died out "centuries" before Göbekli Tepe. Revised dating (Sutikna et al., Nature, 2016) puts its skeletal remains at 100,000 to 60,000 years ago and its tools at about 50,000 years ago.
- The Japanese macaque island where Imo washed sweet potatoes in 1953 is Koshima, not "Kojima."
- Çatalhöyük residents did farm crops and keep sheep and goats, though wild resources remained important.
The chapter then turns to the social nature of discovery. Mlodinow's own thesis work on hydrogen in neutron-star magnetic fields led, through a postdoc's idea about quarks with infinitely many "colours," to a method of solving problems in infinite dimensions, later renamed "dimensional scaling" by Dudley Herschbach. Watt improved a Newcomen engine that had been in use for about 50 years; Newton needed Halley's visit; Einstein needed Marcel Grossmann's mathematics.
Mlodinow's favourite example of animal culture is Imo, a young Japanese macaque who in 1953 began washing sandy sweet potatoes in a stream. Within a few years most of her troop did the same, and later they washed them in the sea, which also salted them. But the macaques never built on the trick. Human culture is different in kind: it accumulates.
Chimpanzees have culture (one 1999 survey listed 38 traditions, from nut-cracking to rain dances), but only humans ratchet: each generation starts from the last one's best tools. What drives the ratchet is not raw intelligence but connection. Google's long cafeteria tables and Bell Labs' corridors (birthplace of the transistor and the laser) were designed for chance encounters. Geneticist Mark Thomas: "It's not how smart you are. It's how well-connected you are."
Before surgery at 76, his father told him he hoped the laws of physics were the same everywhere, and took comfort in being made of the same stuff as "the perfect and romantic stars." The chapter's point is that such a thought requires a culture, a shared, cumulative body of ideas that no single mind could build.
Monuments before farms
The Neolithic turn, in Mlodinow's telling, begins in the mind. Self-awareness (we die) and big questions (who runs the sky?) show up in sites that predate full farming. Göbekli Tepe, about 11,500 years ago in southeastern Turkey, raised T-shaped pillars up to about 18 feet and 16 tons, carved with animals, while its builders still hunted and gathered. Klaus Schmidt, who excavated until his death in 2014, read it as a sanctuary without houses. Later work (see Update) finds domestic buildings, hearths, grinding stones, cisterns and rainwater channels among the monuments. Lee Clare's team at the German Archaeological Institute now describes a settlement with a strong ritual component. In 2026 Necmi Karul's Taş Tepeler work exposed more rectangular second-phase buildings north of the famous enclosures; residential use is the first impression, pending floor-sediment tests. Clare's 2026 palaeoclimate paper ties densest occupation (about 8700 to 8200 calBC) to climate amelioration and later disruption to rapid climate change, slope slips and possible quakes. Either way, communal building at scale arrives before the plough.
Çatalhöyük (about 7500 B.C.) packed up to about 8,000 people without the specialised professions Mlodinow requires for a true city. Bull murals and under-floor burials mark a new stance toward animals and death. Domestication of sheep, goats, cattle and pigs unfolds over roughly two thousand years. Sequence thesis: mental change first, dietary change as follow-through.
- Someone invents a trick (Imo washes a sweet potato, 1953, Koshima).
- Peers copy it.
- Only humans reliably improve the trick and teach the improvement.
- Density and contact raise the rate (Bell Labs corridors; long cafeteria tables). Geneticist Mark Thomas's line, as the book uses it: connection can matter more than raw smarts.
The social nature of discovery is not a soft add-on. Mlodinow's thesis work on hydrogen in neutron-star fields became useful when a postdoc suggested infinitely many quark "colours," later feeding dimensional scaling. Watt improved Newcomen; Newton needed Halley; Einstein needed Grossmann. Imo the macaque washed potatoes; her troop copied; they never built a lab. Human culture ratchets.
A platform team that only copies vendor defaults has macaque culture. A platform team that versions interfaces, publishes migration guides and retires bad patterns has a ratchet. Measure improvements that reduce toil, not only tickets closed.
| Copy culture | Compound culture | ||
|---|---|---|---|
| Unit of progress | The clever individual | The improved practice | Failure mode |
| Fashion, drift | Dogma if critique dies | Design cue | Heroics |
Which shared practice has no named owner for the next improvement? Chapter 3's father story (comfort that the laws are the same everywhere) only makes sense inside a culture that already teaches those laws.
Ratchet mechanics and Neolithic mind
Schmidt's sanctuary reading made Göbekli Tepe famous; the settlement reading makes it harder and more interesting. If people lived among the pillars, then ritual and domestic life were not separate departments. That fits Mlodinow's mental-first thesis without requiring a pure temple on an empty hill. Clare's climate framing adds another upright moral: big projects sit inside environmental constraints, not outside them.
Imo's potato washing is the book's sharpest animal analogy because it fails in a specific way. The troop copies; the trick spreads; no one publishes a better wash. Human science is Imo plus critique plus memory plus strangers who were not there. Bell Labs corridors and Google's long tables are architectural bets on that formula. So are RFCs and design reviews, when they are real.
| Metric | Why it tracks the ratchet |
|---|---|
| Cross-team RFCs merged | Ideas moving past local fashion |
| Mean time to improve a runbook | Critique becoming tool |
| % onboarding steps with named owner | Memory that outlives a hero |
| Chance-encounter hours per week | Corridor substitute |
Churches, parties and firms ratchet too. The difference science claims is built-in hostility to its own last answer. If your ratchet only adds process and never deletes beliefs, you have Imo's cousin: busy, social and stuck.
A control library that only grows becomes phlogiston with checkboxes. Schedule retirement reviews: which control has not prevented an incident in three years and lacks a theory of harm? Delete or rewrite. That is culture with a kill switch.
Culture as ratchet and monumental minds
Göbekli Tepe, Çatalhöyük: coordination, teaching, shared attention. Ratchet clicks when skills transfer; slips when apprenticeship dies.
Karul/Clare and others debate residential versus ceremonial early phases. Large coordinated builds still imply cultural cognition beyond one genius season.
RATCHET = TEACH + TOOL.
Do not back-project Newton onto Neolithic stone. Ruins evidence coordination, not modern method.
Göbekli, Çatalhöyük and the ratchet
The Neolithic is a mental shift as much as a farming one. Göbekli Tepe's monumental enclosures and Çatalhöyük's dense living show culture storing projects across generations. The cultural ratchet: small improvements stick because teaching and imitation outlive inventors.
Work continues under Necmi Karul with Lee Clare coordinating DAI research. Northern exposures show later-phase rectangular buildings read preliminarily as dwellings; floor samples are still under test. Conservation includes Enclosure D. Geophysics points to further unexcavated enclosure space. The site looks less like a pure temple island and more like monument-plus-residence complexity. Mlodinow's "culture before full farming narratives" framing still holds; the residential evidence is the live debate.
Imo washed potatoes; we build on Imo. Macaques copy; humans ratchet. Culture is not cleverness alone: it is cleverness plus connection plus memory that outlives one mind.
Map one practice your team uses that no single hire invented (an onboarding script, a runbook, a review checklist). Trace who last improved it. Add one improvement this month and name the next owner.
A ratchet that only copies local fashion hardens dogma. Göbekli's pillars and Imo's potato wash both show culture; only cumulative critique turns culture into science.
Chapter 4. Civilization
"If I have seen further it is by standing on the shoulders of giants," wrote Newton to Hooke in 1676, borrowing a line already used by Burton and others. The giants of this chapter are anonymous: the people of the first cities.
Uruk, in southern Mesopotamia (near modern Basra, Iraq), grew from about 4000 B.C. to 50,000 to 100,000 people, ten times Çatalhöyük, on the back of canals up to 75 feet wide, the seed plough, barley, wheat, dates and beer. A city meant specialised professions: bakers, brewers, tavern keepers (some women), smelters, weavers and mass-produced bevelled-rim bowls. Centralisation created the intellectual professions too: administration of exchange and storage, armies, public works and formal ways to settle disputes between strangers. Church and state were one: each city was a god's home and the temple ran the economy.
About 85% of early Uruk tablets are accounts: one temple listed 18 bakers, 31 brewers, 7 slaves and a blacksmith. Most of the rest trained accountants. More than 1,000 pictograms (a woman plus a mountain meant a slave woman, foreign women having been brought from the hills) were cut to about 500 signs by 2900 B.C. via the rebus principle, then became syllabic cuneiform. Writing was invented independently only a few times (Sumer before 3000 B.C., Mesoamerica, probably China). The Phoenician alphabet (by 1200 B.C.) fed Aramaic, Hebrew, Arabic and Greek.
Inventors rarely foresee uses: Edison thought the phonograph would record deathbed words. Writing became literature: "tablet houses" (schools) around 2500 B.C., proverbs ("Don't marry a prostitute") and the oldest love poem, about 2000 B.C.
The social history of writing matters as much as the technique. Scribes formed a literate class; in Egypt they were exempt from taxes and manual labour, and a well-known text urges a student to become a scribe because every other trade is miserable. Mlodinow recalls a Cub Scout assignment, at eight or nine, to invent a writing system. His looked strikingly novel until the pack leader asked whether it was based on the Hebrew alphabet; it was. The children had an advantage the first inventors lacked: they had already been taught that speech breaks into elementary sounds that can be mapped onto letters. And writing was not a natural extension of speech. Language is innate (every human group speaks, and the linguist Steven Pinker notes there is no mute tribe; in Aboriginal Australia some 500 groups of about 500 people each had their own language), but writing had to be invented and was invented only a handful of times.
| Speech | Writing | ||
|---|---|---|---|
| Origin | Evolved, innate; every human group has it | Invented, a few times only (Sumer, Mesoamerica, probably China, perhaps Egypt) | Learning |
| Absorbed by children without instruction | Requires years of schooling | First use | Social life |
| Accounting and administration | Effect on knowledge | Memory limited to individuals | Knowledge stored, copied and checked across generations |
Mathematics began the same way. Babylonian scribes had recipes for quadratic and even cubic equations and estimated canal labour, but stated them in prose ("4 is the length and 5 the diagonal..."). Mlodinow calls notation the unsung hero. Indian mathematicians around A.D. 500 gave us base ten with zero as a number, negative numbers (debts) and symbols for unknowns; Europe got plus and minus signs in the fifteenth century and the equals sign from Robert Recorde in 1557. Egypt, whose Nile flooded for four months a year, made geometry ("earth measurement") a practical art: thirteenth-century-B.C. builders levelled a 50-foot beam to within 1/50 inch.
- In 2017 Oxford's Bodleian Libraries announced radiocarbon dates placing the Bakhshali manuscript, with its dot for zero, as early as the third or fourth century A.D. A 2024 re-dating by Oxford researchers revised this to roughly the eighth to eleventh centuries, so the "oldest zero" headline should not be repeated.
- The book's ~3,000 spoken languages is far too low; standard catalogues list about 7,000.
Numbers are partly innate: infants can tell sets of up to about four objects apart without counting. Everything beyond is cultural technology. Saint Augustine, around A.D. 415, warned Christians against "mathematicians" who had made "a covenant with the devil," but he meant astrologers, then called mathematici. The confusion shows how closely numbers were tied to divination before they became a tool of science.
The final invention is law. Mesopotamian gods numbered in the thousands (a god of brewing, a demon called Fever), and religious codes included rules like "Don't vomit in streams." The Code of Hammurabi (about 1750 B.C.), carved on an 8-foot basalt stele found in 1901 and now in the Louvre, was handed down by the god Marduk, who also legislated to the stars. The code was no charter of rights. The upper classes and royals received leniency, and the penalties were harsh: a robber should be put to death; anyone who stole while helping to put out a fire should be thrown into the fire; a "sister of god" (a priestess) who opened a tavern should be burned; a man whose dam failed because he was "too lazy" to maintain it had to replace his neighbours' ruined corn. Its importance lies elsewhere: it shows law as a written, public, general rule, applied case by case, which is the template for natural law.
From such codes the Greeks borrowed the idea of nature obeying law: Anaximander's elements "pay fine and penalty"; Heraclitus's sun will not "overstep his measures"; astronomy contains nomos, law. Even Kepler gave planets "minds." As the historian Edgar Zilsel put it, humans "interpret nature after the pattern of society."
In September 1939 German troops entered Częstochowa, where Mlodinow's father ran a tailoring shop. Within days decrees confiscated Jewish property and businesses. Human law could be overturned overnight. The laws of physics cannot, and that, the son argues, is part of their comfort.
Tablets, numbers, law
Uruk grew from about 4000 B.C. toward 50,000 to 100,000 people on canals, ploughs and barley. Specialised work (bakers, brewers, smelters, weavers) and temple administration created demand for memory outside the skull. Writing answered accounting first: the book cites about 85% of early tablets as accounts. Pictograms numbered over 1,000, then shrank toward 500 signs by about 2900 B.C. as the rebus principle turned pictures into sounds. Literature and love poems came later. Speech is universal; writing is rare. Inventors rarely foresee uses: Edison thought the phonograph would record deathbed words; Sumerian accountants accidentally unlocked literature.
Mathematics rode the same wave. Babylonian scribes solved quadratics in prose recipes. Indian mathematicians around A.D. 500 gave base ten with zero as a number; Europe accepted plus and minus centuries later; Robert Recorde's equals sign arrives in 1557. Notation is not cosmetics. It is compressed reasoning you can check. Infants can tell small sets apart without counting; everything beyond about four is cultural technology.
| Tool | First job | Later gift to science | |
|---|---|---|---|
| Writing | Accounts and administration | Cumulative, checkable knowledge | Mathematics |
| Labour, land, trade | Prediction by calculation | Law | Dispute settlement among strangers |
Law completes the trio. The Code of Hammurabi (about 1750 B.C.), on an 8-foot basalt stele now in the Louvre, is harsh and class-biased, yet it is written, public and general. Greek thinkers later borrowed the pattern for nature: elements that "pay penalty," a sun that will not overstep. Edgar Zilsel's point, as the book uses it: we read nature through social forms we already know. Human law in Częstochowa in 1939 could flip overnight; physics could not.
The path the book draws is short to state and long to walk: account → number → law → nature as law. Skip a step and you get astrology dressed as mathematics, or metrics without governance.
A credit policy that lives in a slide deck is Homeric custom. A credit policy with versioned rules, published exceptions and an audit trail is closer to a stele. Keep a physics-like layer: invariant checks and reproducible scores.
Mlodinow's Cub Scout story (reinventing an alphabet that turned out to be Hebrew) underlines how hard the first invention was. Children who already know that speech breaks into sounds are not starting from the Sumerian blank page. Language is innate; writing is a technology.
Administration invents abstraction
The scandal of writing is its boring origin. About 85% accounting is not a footnote; it is the plot. Abstraction hitchhiked on bureaucracy. That should humble theorists and embolden operators: your ledger may be the next alphabet.
The rebus principle is the chapter's hidden algorithm: reuse a sign for its sound, not its picture. Many modern abstractions work the same way. A "shopping cart" icon no longer pictures furniture; it sounds like checkout. APIs begin as pictures of internal tables and become sounds other services speak.
| Dimension | Language | Writing | |
|---|---|---|---|
| Presence | Every human group | Invented a few times | Acquisition |
| Childhood absorption | Years of schooling | First function | Social life |
| Accounts | Failure mode | Misunderstanding | Fraud, loss, silence |
Hammurabi's harshness is the point, not an embarrassment to skip. Public general rules can be unjust and still be a template for natural law thinking. Greeks borrowed the template; they did not borrow every penalty. When you design policy engines, separate "is written and testable" from "is good."
Robert Recorde's = (1557) is a UX win for reasoning. What symbol does your team lack? A standard way to write "this metric equals that definition across services" prevents months of prose recipes that hide bugs.
Writing, law and public rules
Uruk, tokens, tablets, notation, Hammurabi: externalised cognition. Science borrows notebooks, equations, standards.
Public rules, private skill
| Layer | Example | Science borrow | |
|---|---|---|---|
| Token / tablet | inventory | lab notebook | Notation |
| numbers | equations | Law | public rules |
| methods | Profession | scribe | specialist roles |
Uruk, tokens and the law tablet
Newton's line about standing on the shoulders of giants (1676 letter to Hooke) opens this chapter for a reason. Giants need scaffolding: cities, surplus, professions, writing and law. Without them there is nowhere for a Galileo or a Newton to stand.
Uruk was the New York of its day. Inhabitants used a seed plow that dropped grain into a furrow while it cut, a specialized tool that demanded training. Specialization is the civilizational signature: not everyone farms, so someone can scribe, weigh or argue. Larger groups force pretence and restraint. Smashing a clay tablet on a neighbour's head becomes costly when you will see him tomorrow. That soft coercion is the social half of writing's hard technology.
Inventors of new recording systems often do not grasp what they have made. Early tokens and seals tracked goods; full language writing arrived later. Treat your first database schema the same way: it is inventory before literature.
Before the Neolithic shift, languages may have been nearly as numerous as tribes. Australia before late-eighteenth-century British colonization is the book's ethnographic pointer. Shared script collapses that fragmentation into a ratchet that can store debts across seasons.
Scribes trained for years on thousands of cuneiform signs. Complexity was a feature and a tax. Egyptian practical geometry (Greek for "earth measurement") leveled a fifty-foot pyramid beam in the thirteenth century B.C. with an error of about one-fiftieth of an inch. That is not Euclid yet. It is grade-school rules that work: arithmetic for grain, geometry for stone.
Hammurabi's code is not modern democracy. Royals got leniency; slaves could be sold or killed. It still installed public, written thresholds: eye for eye, payment for injury, named offences. Torah harshness arrives later on a related line. Law as tablet is civilization's mean-speed theorem: predictable consequences replace whim.
Mlodinow ties the close to his father's shattered world. Law that can be burned or rewritten at a madman's whim is not scaffolding; it is stage set. The Holocaust, he notes, has smaller-scale cousins across history. Civilization is reversible when the tablets lose force.
Vague OKRs are oral custom. Tablets scale.
Missing pieces teams still skip: sexagesimal inheritance in minutes and degrees; clay tokens before full script; cylinder seals as identity; the long path from Egyptian practice to Greek proof. Put them on your civilizational checklist before you claim a "platform."
City gods, Indian zero and reciprocal life
Religion in early cities was executive power, not only belief. Each city had a patron deity thought to dwell there; decline meant the gods had left. Fear of the gods motivated obedience and moved goods through temple economies. Writing then locked reciprocal city life: buying, selling, billing, shipping, borrowing, promising. A week with no written record of any transaction is modern chaos; Uruk lived that risk until script scaled.
Indian mathematicians from around A.D. 500 brought base ten, zero as a number (multiply by zero yields zero; add zero and the number stands), and negatives for debt. Prose equations gave way to notation that could travel. That notational leap is as civilizational as the seed plow. Teams that keep strategy in slide prose and never invent a notation for their domain are stuck pre-India.
Ledgers that cannot represent zero balances and negative positions cleanly recreate Babylonian awkwardness. Your core schema is a civilizational choice.
Writing began as a receipt. About 85% of early Uruk tablets are accounts. Abstraction rode on administration. Your "theory" may first need a ledger.
Take one fuzzy claim on your roadmap ("users love feature X"). Turn it into a tablet: who, what count, what date, what threshold. If you cannot fill those four cells, the claim is not ready to guide spend.
Ledgers without law become bureaucracy. Hammurabi's stele shows written rules can be harsh and uneven. Measure what matters, and keep a path to revise the rule.
Chapter 5. Reason
When Alexander the Great, aged 22, crossed the Hellespont in 334 B.C., Uruk was already ancient, and the Greek world he spread had a new idea: cosmos instead of chaos.
Before the Ionians, the Greek universe was capricious. Homer and Hesiod, written down around 700 B.C., describe a world born from chaos (the Greek word for a yawning void) and run by gods whose quarrels explain storms, plagues and wars. Alexander's conquests later spread Greek ideas east: the Babylonian priest Berossus wrote his people's history in Greek, and taxes in Egypt were recorded in Greek on papyrus. Hellenisation carried both the myths and their rational critique.
In the sixth century B.C., in Ionia (the same era as the Buddha and Confucius), Miletus, a rich trading port the book compares to "an ancient New York," produced Thales. He cornered the olive presses, measured the pyramids by their shadows, is credited with the first proofs in geometry and explained earthquakes as the floating Earth rocking on water. The explanation was wrong; the method was new: a natural cause defended by argument. Aristotle called such thinkers physikoi (students of physis, nature), as opposed to the theologoi.
Why Miletus? The book's answer is economic and social. A port with colonies around the Black Sea, it mixed Greeks, Lydians, Phoenicians and Egyptians, and its wealth created leisure. Contact with foreign customs made local myths look local; leisure made it possible to argue about them. The same combination, diversity plus surplus, reappears in Renaissance Florence, Enlightenment Edinburgh and modern research clusters.
Mlodinow also stresses why mathematics became the language of science. A physicist's equation is compressed reasoning: write down the law, and its consequences unfold by calculation, including consequences no one expected. Thales's proofs were the first hint of that power.
Thales was also, by tradition, a shrewd businessman, an exception to the rule that thinkers needed patrons. He is said to have made a fortune by cornering the market on olive presses and charging high prices for the oil, "like a one-man OPEC," and to have known Miletus's dictator, Thrasybulus, intimately. Wealth bought him time to think. The poet Rilke advised a young correspondent in 1903 to "love the questions themselves"; Thales's achievement was to make the questions answerable.
Pythagoras (about 570 to 490 B.C.) found that a string's pitch rises as its length falls (half the length, an octave higher) and concluded that number is the essence of reality. He probably did not prove the theorem named after him, which Babylonians knew a millennium earlier, but his reverence for number reached Plato and, much later, Galileo.
| Strength | Weakness |
|---|---|
| First to study nature systematically and encyclopedically (about 170 works, a third surviving) | Qualitative, not quantitative: no clocks, no concept of speed as a number |
| Careful observation and dissection in biology | Teleology: things move to fulfil a purpose, which stalled physics for 2,000 years |
| Distinguished natural from violent change | Projectiles kept moving because air rushed in behind them |
| Insisted claims be argued, not asserted | Rejected atoms; treated common sense ("what everyone believes is true") as evidence |
Pythagoras himself is half legend. Born on Samos, he is said to have met the aged Thales, left Samos at about 40 to escape the tyrant Polycrates and founded a semi-religious brotherhood at Croton in southern Italy. A later story has him passing a blacksmith's shop and noticing that hammers of different weights rang in harmony, which led him to the mathematics of pitch. Whatever the truth, the historian Carl Huffman notes that the Pythagoreans gave number a dignity it had never had: reality itself was taken to be mathematical.
Aristotle's background suited a natural historian. Born in Stagira in northeastern Greece, the son of the physician to Alexander's grandfather, King Amyntas, he was orphaned young and sent to Plato's Academy at seventeen, where he stayed twenty years. After tutoring Alexander he returned to Athens and founded the Lyceum, teaching while walking in its gardens (hence "peripatetic").
Aristotle (384 to 322 B.C.) studied at Plato's Academy for 20 years, tutored Alexander and founded the Lyceum. Bertrand Russell called him the first to "write like a professor"; Daniel Boorstin warned of his "tyranny of common sense." His physics described four elements with natural places and said force was needed to keep things moving. Mlodinow's critique is that he never quantified: an apple falling gains 22 mph each second, a fact you can only state with numbers and clocks. And teleology, the idea that a stone falls because it wants to reach its place, made questions about mechanism seem unnecessary.
Mlodinow contrasts Aristotle's purpose-driven world with the indifferent laws of modern physics, using a painful story: his mother, at 17, watching a Nazi shoot Jews kneeling in the snow. His father preferred a universe of impartial laws to one where such things had a purpose. Analytically, the shift from "what is it for?" to "what rule does it follow?" is what made prediction possible.
Mlodinow's critique of quantification has a vivid form. Speed as a number, such as 60 miles per hour or 88 feet per second, had no meaning for Aristotle, who had no clock that measured seconds and no concept of a ratio of distance to time. Without such numbers you cannot even state that a falling apple gains 22 miles per hour every second, let alone discover it. He also notes a paradox of Aristotle's force: on his principles, pushing a chair across a room should require a force comparable to the chair's weight, and stopping pushing should stop it at once. Everyday experience with carts and ships seemed to agree, which is why it took so long to see the friction hidden in the observation.
Mlodinow knew a physics professor who made lunch from the cafeteria's free condiments table (mayonnaise for fat, ketchup as his vegetable, saltines for carbs) and a friend who ordered a plate of salami and ate it with knife and fork like a steak. The stereotype of the odd scientist has some truth, and he argues it is a feature: "science is the natural enemy of preconceived notions, and of authority." The biggest barrier to progress is "an undue allegiance to the ideas of the past, and present," so if he were hiring for innovation he would look for people willing to fly in the face of what everyone believes. Aristotle's 2,000-year dominance is the cautionary opposite.
His explanation of projectiles shows the system's weakness. Since nature "abhors" a vacuum, he said, air rushes in behind a thrown javelin and pushes it along. (Mlodinow notes that Tokyo subway attendants seem to apply the same principle when packing passengers into cars.) It was an ingenious rescue of a theory in trouble, and a classic case of adding an assumption to protect a framework rather than testing it.
Aristotle fled Athens after Alexander's death so the city would not "sin twice against philosophy," and died in 322 B.C. His works were revived through Arab philosophers and absorbed into Church teaching, so Europe studied Aristotle for nineteen centuries. Fourteenth-century cannons finally made his projectile theory absurd: soldiers needed trajectories.
- Science rests on old tools: curiosity, culture, writing, mathematics, law and argument.
- Connection beats genius: ideas ratchet forward in dense, linked communities.
- Writing and maths began as administration; abstraction came later.
- The Greeks replaced gods with natural causes, but Aristotle's qualitative, purpose-driven physics shows that reason without measurement can stall for centuries.
- Since 2015, our species has aged by 100,000 years and Göbekli Tepe has gained houses, but the argument holds.
Cosmos instead of chaos
Alexander crosses the Hellespont in 334 B.C. carrying a Greek idea already older than he is: the world as cosmos, an ordered whole, not a playground of temperamental gods. If the United States had existed as long as Uruk, Mlodinow notes, we would be on roughly our six hundredth president. The Near East was not in decline; Greek-speaking culture had simply surpassed it intellectually, and Alexander, schooled by Aristotle, carried that culture east.
Thales of Miletus, in the sixth century B.C., offers natural explanations and geometric proofs. Wrong on earthquakes (Earth floats on water); right on method: a natural cause defended by argument. Pythagoras hears number in vibrating strings and elevates number into the stuff of reality. Aristotle builds the first encyclopedic natural history and then freezes physics with teleology and qualitative motion for roughly two thousand years. Why Miletus? Trade, mixing of peoples and surplus leisure. Local myths look less inevitable when foreign myths sit at the next table.
Gave: systematic observation, biology, the demand that claims be argued. Blocked: speed as a number, force laws without purpose, atoms as serious ontology. Cannons in the fourteenth century made his projectile story look absurd; Galileo finished the job.
If a candidate explains a system only by what it is "for," press for the rule it follows when purpose fails. Aristotle hired well for natural history and poorly for dynamics.
The olive-press story (Thales cornering presses ahead of a good harvest) reminds that leisure for thought often rides on surplus. Aristotle's projectile rescue (air rushing in behind the javelin) is ingenuity spent defending a framework rather than testing it. Mlodinow's mother, at 17, watching a Nazi shoot Jews in the snow, frames why his father preferred impartial laws to a cosmos with a purpose for such things.
Large models invite purpose-talk ("the model wants to help"). Rewrite: the model minimises a loss on a distribution. That is law-talk. It keeps you honest when the model harms.
| Rung | Achievement | Failure mode if stuck | |
|---|---|---|---|
| Curiosity | Ask why | Freeze | Culture |
| Ratchet ideas | Fashion without critique | Civilization | Write, count, legislate |
| Bureaucracy | Reason | Natural causes + maths | Teleology without measure |
If you cannot name the mechanism, label it preference, not inference.
Teleology's long shadow
Thales is the book's patron saint of wrong-but-methodical. Celebrate him without repeating his geology. Pythagoras is the patron of maths as ontology; useful until it becomes numerology. Aristotle is the patron of encyclopedic care and dynamical failure. Nineteen centuries of curriculum power show that being systematic is not the same as being quantitative.
Mlodinow's force paradox (pushing a chair should feel like lifting its weight if Aristotle were right) is a classroom for friction. Everyday experience concealed the variable that mattered. Cannons made the concealment untenable. Your legacy systems have cannons: load tests, incident reviews, customer escapes. Use them.
The condiment-lunch professor and the salami-as-steak friend are comic. The serious claim: science needs people willing to fly in the face of what everyone believes, plus a community that can still kill bad ideas. Oddness without critique is a cult. Critique without oddness is a committee.
"What everyone believes is true" is a training-data risk. Models absorb teleology-like shortcuts ("loans exist to help families") that hide mechanisms ("loans maximise a constrained objective"). Force mechanism language in model cards.
Part I takeaways deserve a sharper edge: connection beats lone genius; writing and maths began as admin; Greek reason without measurement can stall for centuries; 2015 chronology moved (species age, Göbekli houses) without killing the argument.
Ionia, teleology, unlearning
Thales, Pythagoras, Ionian wager: speakable patterns. Aristotle organises including teleology that physics must partially unlearn.
Unlearning teleology
| Habit | Pays | Misleads | |
|---|---|---|---|
| Pattern without myth | regularities | noise as law | Teleology |
| careful functions | inertia, projectiles | Deduction | geometry |
Ionia, Thales and purpose versus law
Ionia put thinkers where trade met argument. Thales predicted and measured; Pythagoras made number a cosmic claim; Aristotle built a vast qualitative physics with teleology (purpose) in the driver's seat. The chapter's hinge: purpose-talk feels explanatory and blocks law-talk. Galileo will later mathematize motion; Newton will finish the divorce from final causes in mechanics.
For any design doc, highlight every "so that" that names a purpose without a mechanism. Rewrite half of them as force laws, rates or constraints before review.
Cosmos, not chaos. Thales offered a wrong cause (Earth floats on water) with a new method: argue from nature. Wrong answers with right method beat right-sounding myths.
Rewrite one "because leadership wants it" decision as a natural cause claim: what measurable mechanism would make the decision correct? If you cannot name the mechanism, label the decision as preference, not inference.
Aristotle's trap: purpose posing as explanation. "The stone falls to find its place" sounds deep and blocks measurement. Ask "what rule?" before "what for?"
Drive. Scarcity-grade curiosity; two clocks; acceptance lag as plot. Curiosity. Costly brains; infant labs; probe before narrative. Culture. Ratchet; Göbekli-scale coordination; no Newton back-projection. Civilisation. External mind stack; record habits. Reason. Ionian wager; unlearn teleology for motion; let results veto axioms.
| Chapter | Habit |
|---|---|
| 1 Drive | watch acceptance lag |
| 2 Curiosity | probe before narrative |
| 3 Culture | teach so the ratchet clicks |
| 4 Civilisation | write the decision |
| 5 Reason | let results veto axioms |
Residence debates and DESI DR2 evolving-dark-energy hints (~2.8-4.2σ in community summaries): schedule the next measurement; do not flip slogans early.
Part I close. The upright story through reason is externalised curiosity: posture frees hands, culture teaches, civilisation records, reason argues, and acceptance lag decides whether any of it compounds. Keep a private ledger of ideas you conceived versus ideas your organisation accepted this quarter. The gap is the real curriculum.
Part II. The Sciences
"The dogmas of the quiet past are inadequate to the stormy present... As our case is new, so we must think anew, and act anew." Abraham Lincoln, annual message to Congress (1862), the epigraph to Part II
Part II moves from millennia to centuries. Its subject is how Europe, after a long detour, learned to measure nature, test ideas by experiment and write the results in mathematics, and how four disciplines (physics, chemistry, biology and, by implication, the scientific method itself) took modern form.
Chapter 6. A New Way to Reason
Mlodinow opens with a debate he co-wrote with Deepak Chopra (War of the Worldviews). His position is reductionist: in principle, atoms and four forces explain everything, even if no equation will ever describe love. The question for this chapter is how the West came to believe that.
The long detour
Rome conquered Greece (146 B.C.) but, as Cicero admitted, limited geometry to "measuring and counting"; there was no Roman mathematician of note. After 476 the Western cities shrank and Greek learning survived in monasteries, mostly forgotten. The Islamic world translated Greek works and advanced optics, astronomy, mathematics and medicine, then, in the book's account, declined from the thirteenth century under a narrow doctrine of utility, Mongol invasion and religious schools (madrassas) that excluded natural science. China, with over 100 million people in 1200 to 1500 (about twice Europe's population), had a rigid examination system; India had caste.
The claim that Islamic science declined after the thirteenth century, and that China and India produced nothing comparable, is contested by historians. George Saliba and others point to major later work (the Maragha and Samarkand observatories, Ibn al-Shatir's planetary models in the 1300s, which closely resemble Copernicus's), and Joseph Needham's volumes document centuries of Chinese invention. The "Needham question" (why the scientific revolution happened in Europe) is still debated; treat the book's answer as one view.
The revival came through translation: Constantinus Africanus rendered Arabic medical texts into Latin in the late eleventh century, and the reconquest of Toledo in 1085 opened its libraries. Patrons sponsored translations as status symbols. Then came the universities, organised like guilds: Bologna (1088), Paris (about 1200), Padua (about 1222), Oxford (by 1250). Student life was rough: a 1495 German statute banned drenching freshmen in urine, and Bologna students fined professors who started late.
Professors had it no easier. Many had no classroom and lectured in rooming houses, churches, even brothels, and were often paid directly by students, who could hire and fire them. Yet these rowdy guilds became the first permanent institutions in which natural philosophy was a career, with libraries, curricula and the habit of disputation.
At a party, Mlodinow praised medieval scientists and a guest scoffed: they operated without anaesthetic, made potions of lettuce juice, hemlock and the gall of a castrated boar, and even Thomas Aquinas believed in witches. He checked later; she was right. His reply is that the same era produced the Merton scholars and Oresme, which makes their achievement "all the more miraculous" given the state of medieval knowledge, the cold, the hunger and the Black Death of 1347.
The Merton calculators and Oresme's graph
The "scientific revolution" (roughly 1550 to 1700) had medieval foundations. At Merton College, Oxford, between 1325 and 1359, a group of scholars singled out motion as the most fundamental kind of change and studied it without asking its purpose. That was hard in a world where time was subjective: the first clocks striking equal hours appeared in the 1330s, and a London "hour" varied from 38 to 82 minutes across the seasons.
A body accelerating uniformly from rest covers the same distance as a body moving all the time at its average speed. A car going from 0 to 100 mph in a minute travels as far as one cruising at 50 mph for that minute. In modern notation: distance = ½ × a × t² = ½ × (a × t) × t.
The rule was proved geometrically by Nicole Oresme (1320 to 1382) in Paris, later bishop of Lisieux, using what was in effect the first graph: speed plotted against time, with distance as the area underneath. Galileo later wrote that the book of nature "is written in the language of mathematics"; Oresme was among the first to read it.
Oresme's proof illustrates a recurring pattern: to advance, physicists often have to invent new mathematics. If mathematics is the language of physics, the lack of the right mathematics leaves a physicist unable to speak, or even reason, about a topic. Einstein needed his friend Grossmann's differential geometry for general relativity; Newton needed calculus, which he invented. Oresme did not publish widely and his ideas were not taken up for two centuries, but the graph as a tool for thinking about change is his.
Engines, printing and a plural Europe
Late medieval Europe became the first civilisation powered largely by waterwheels and windmills rather than muscle. Printing (movable type in China about 1040, Europe about 1450) changed the economics of ideas: Florence's Ripoli Press in 1483 charged three times a scribe's fee to set up a book but then produced over 1,000 copies; by 1600 about 1,000 mathematics texts had been printed. Europe was plural (town and country, Church and state, pope and emperor), which left room for dissent. After the Black Death killed a third to a half of Europeans, Renaissance art brought anatomy and perspective; Leonardo left more than 10,000 pages of notes.
Galileo
Galileo was born in Pisa in 1564, two months before Shakespeare. He dropped out of medicine, lost a Bologna post at 23 (after claiming to be "around twenty-six") and taught astrology to medical students at Pisa for 60 scudi a year before a satirical poem against the academic gown helped him leave for Padua at 180 scudi, his "best eighteen years."
Galileo's money worries shaped his early career. As a Pisa lecturer earning 60 scudi a year, he charged students twelve scudi for an astrological reading; five readings a year doubled his salary. He also gambled, and in an era before the mathematics of probability he was both a pioneer in calculating odds and a good bluffer. His later fame should not hide the fact that he spent most of his life seeking patrons and income.
His method was exploratory and quantitative. Aristotle said heavy bodies fall faster. Galileo slowed falling by rolling bronze balls down inclined planes, timed them with a water clock (by weighing the water that flowed out) and found that distance grows with the square of time, independent of weight. Tilting the plane toward vertical took him, by extrapolation, to free fall, and to the conclusion that in a vacuum all bodies fall equally. He accepted idealisation: to see the law, you must "cut loose from these difficulties" of friction and air.
| Time (units) | 1 | 2 | 3 | 4 |
|---|---|---|---|---|
| Total distance | 1 | 4 | 9 | 16 |
| Distance in that interval | 1 | 3 | 5 | 7 |
The odd-number pattern is the Merton rule made visible: the speed grows steadily, so each interval adds two more units than the last.
Aristotle's claim that heavier bodies fall faster had always violated common sense in one respect. As the Jesuit astronomer Giovanni Riccioli pointed out, even the mythological eagle that killed Aeschylus by dropping a turtle on his head knew that a falling object does more damage from higher up, which implies that falling objects speed up. Scholars had debated the point for centuries; Galileo was the one who measured it.
| Question | Aristotle | Galileo |
|---|---|---|
| Why do things fall? | To reach their natural place | Not explained (no force of gravity yet) |
| Does weight change fall speed? | Yes, heavier falls faster | No, apart from air resistance |
| Does falling speed change? | Vaguely, near the end | Grows steadily; distance grows as time squared |
| What keeps a moving body moving? | A continuing force (or air pushing behind) | Nothing: motion persists unless resisted (inertia) |
| Method | Observation and argument | Measurement, idealisation, thought experiment |
His thought experiment of a ship's cabin (butterflies, fish and dripping water behave the same whether the ship is moored or moving smoothly) led to the law of inertia. Mlodinow's father called Galileo the Moses of science: he led the way but did not reach the promised land, since he kept circular natural motion and had no law of gravity.
The telescope turned Galileo from a professor into a celebrity. Hearing in 1609 of a Dutch spyglass of two or three powers, he built a nine-power instrument within months; his friend Paolo Sarpi saw that unwalled Venice could use it to spot ships hours earlier. Turning it on the Moon in December 1609, he watched shadows move across the terminator over weeks and, using the Pythagorean theorem, estimated the heights of lunar mountains. The Moon was a rough world, not a perfect sphere. The four moons of Jupiter, which he named the Medicean stars after the Grand Duke's family, showed that not everything circles the Earth. The Starry Messenger, about 60 pages, sold out.
| Year | Event |
|---|---|
| 1543 | Copernicus publishes On the Revolutions, dedicated to the pope, on his deathbed |
| 1609 | Galileo builds a 9-power telescope in months (later 20 to 30) and gives it to the Venetian Senate; salary doubled to 1,000 scudi |
| 1610 | The Starry Messenger: mountains on the Moon, four moons of Jupiter; becomes court philosopher to Cosimo II in Florence |
| 1616 | Ordered not to hold or teach heliocentrism |
| 1632 | Dialogue, in which the pope's argument is voiced by Simplicio |
| 1633 | Convicted and forced to abjure (June 22); house arrest at Arcetri |
| 1638 | Two New Sciences smuggled to Leiden |
| 1642 | Dies, aged 77, the year Newton is born |
The telescope's origin was almost accidental. Two children playing in the shop of a Dutch spectacle maker, Hans Lippershey, put two lenses together and saw a distant weathervane magnified. When Galileo heard of it in 1609 he was unimpressed. He became interested only because his friend Paolo Sarpi, described by the historian J. L. Heilbron as a "redoubtable anti-Jesuit polymathic monk," saw military value for Venice, an unwalled city that depended on early warning of attack.
In the Dialogue, Galileo had told Urban VIII that his purpose was to defend the Church from the charge that it had banned heliocentrism out of ignorance. The book took the form of a conversation among Simplicio, a follower of Aristotle, Sagredo, an intelligent neutral, and Salviati, who argued for Copernicus. Instead of stating the pope's disclaimer in his own voice, as Urban had asked, Galileo put it in the mouth of Simplicio, whom Heilbron calls a "nincompoop." Urban, "being no nincompoop himself," was deeply offended.
The trial, Mlodinow argues, was about who owns truth. Galileo's friend Maffeo Barberini, as Pope Urban VIII, felt mocked by the Dialogue, and the Giordano Bruno precedent (burned in 1600) was recent.
Galileo's abjuration on 22 June 1633 required him, "aged seventy years," to "abjure, curse and detest" his errors; Giordano Bruno had been burned in Rome in 1600 for similar views. Two days later he was released to the Florentine ambassador and then confined to his villa at Arcetri, where he could receive visitors, even heretics, provided they were not mathematicians. One was the young John Milton, who later mentioned Galileo's telescope in Paradise Lost. Under house arrest Galileo wrote his greatest work, Two New Sciences, on motion and the strength of materials; it had to be smuggled to Leiden for publication in 1638. When he died in 1642 the pope blocked a monument, but the librarian of Cardinal Barberini's court wrote that his death "touches not just Florence, but the whole world."
The book says a daughter of Galileo died of plague far away. His beloved elder daughter, Sister Maria Celeste, died in 1634 at the convent of San Matteo in Arcetri, near his house, soon after his return from Rome.
- In August 2018 historian Salvatore Ricciardo found Galileo's original 1613 letter to Benedetto Castelli in the Royal Society's archive. Its edits show Galileo himself softened his wording on Scripture after copies circulated, rather than enemies having doctored it.
- The Church's formal reckoning came on 31 October 1992, when John Paul II closed a commission opened in 1981 and acknowledged that theologians had erred.
- Galileo found four moons of Jupiter. The confirmed count reached 115 by April 2026, after new discoveries confirmed by the IAU Minor Planet Center.
A new way to reason: from Merton to Galileo
Mlodinow frames the leap from Greek philosophy to mathematical physics with a personal contrast: collaborating with Stephen Hawking and with Deepak Chopra, whose worldviews barely overlap. Among scientists, the mathematical approach to the physical world is now nearly universal. Getting there required a long detour. Rome conquered Greece in 146 B.C.; much of the Greek toolkit slept, traveled through Arabic scholarship, and re-entered Europe incompletely.
Medieval Europe was not empty. The Merton calculators at Oxford and Nicole Oresme in Paris developed the mean speed theorem: a uniformly accelerated body travels as far as one moving at its average speed. Oresme even graphed it. The idea lived on parchment before it lived in a lab. Engines, printing and a plural Europe (no single empire able to throttle every press) then lowered the cost of arguing in public.
Distance under constant acceleration equals distance at the average of initial and final speeds. Galileo's inclined-plane numbers (1, 4, 9, 16) make the odd-number increments (1, 3, 5, 7) visible: speed grows steadily with time.
Galileo (1564 to 1642) put the rule on a ramp. He timed balls, found distance growing as the square of time, and replaced Aristotle's qualitative motion with numbers. He aimed a telescope at the sky and found moons of Jupiter, phases of Venus and a Moon that was not a perfect sphere. The Church fight was real; so was the method: measure, mathematize, publish.
| Aristotle | Galileo | |
|---|---|---|
| Falling bodies | Heavier falls faster | Same acceleration (in vacuum) |
| Method | Qualitative purpose | Timed experiment + maths |
| Projectiles | Air pushes from behind | Composition of motions |
| Authority | Texts and common sense | Observation that can surprise |
Mlodinow recalls being challenged on whether science is just another belief system. The upright answer is procedural: science is the set of claims that survive hostile measurement. Your architecture review can borrow the test: which claims in this design have been measured by someone who wanted them to fail?
If a process is only described with adjectives ("slow", "spiky", "healthy"), you are closer to Aristotle than to Galileo. Instrument one quantity for a week. Keep the plot; retire one adjective.
Link forward to Newton, who inherited a world already half-mathematized.
Rome, parchment and the ramp
The leap from Thales to Galileo was blocked first by Rome. Cicero praised Greek honour for geometers while boasting that Romans confined maths to the useful. Culture can starve a method even when the texts still exist. After the Western Empire's dissolution, Greek science traveled through Arabic scholarship and re-entered Europe unevenly. Interest in philosophy and maths declined among elites who prized practice over theory. That is Chapter 3's ratchet run in reverse.
Mlodinow's Chopra contrast is not a digression. Calling mathematical physics reductionist as an insult only works in rooms where equations feel soul-less. The chapter's wager: the same method that felt foreign after Rome became the default for matter and motion.
Merton and Oresme show theory can lead. Galileo shows apparatus decides. Teams that only write design docs are Merton without a ramp. Teams that only ship A/B tests without a mean-speed theorem are ramps without a curriculum.
"Only build what users ask" is a Roman filter. It kills geometers. Keep a budgeted non-utilitarian probe stream (5 to 10% time) with kill criteria, so you are not Cicero with a backlog.
Urban VIII, the telescope's moons and phases, and house arrest are the acceptance drama. Hold the method: measure, mathematize, publish under constraint.
Merton, Oresme, print, Galileo; diagrams as arguments
Universities, mean-speed rule, graphs, printing, Galileo’s maths-plus-bench. The new way is social: a community that accepts losing to a result. Authority preferring book over bench creates acceptance lag with teeth.
Merton, Oresme and the Galilean ramp
Medieval universities kept Aristotle alive while Merton Calculators stated the mean-speed theorem and Oresme drew change on a graph. Printing spread the arguments. Galileo's ramp and telescope turned parchment into apparatus. Moons of Jupiter and phases of Venus made the old cosmology negotiate with light. House arrest shows acceptance politics: conception can win in notebooks while authority still owns the street.
Count theorems written and count apparatus hours separately. Either alone is a stalled Chapter 6.
Measure the fall. Galileo timed balls on a ramp and found distance grows as the square of time. The Merton rule left the page and entered the lab. Theory dies or lives by numbers.
Pick a process you only describe in adjectives ("slow", "spiky"). Instrument it for one week so you can plot a quantity against time. Keep the plot; throw out one adjective.
Quantifying the wrong variable is false precision. Galileo's Church fight also warns: institutions punish methods that threaten their story even when the data are clean. Pair every new metric with a kill criterion and a named rival metric so precision cannot become theatre.
Chapter 7. The Mechanical Universe
After Newton, the universe looked like a clock: the Declaration of Independence appealed to "the Laws of Nature," and Laplace imagined an intellect that, knowing every force and position, could see the future like the past. We still think like Newtonians when we speak of momentum and force of character.
A difficult genius
Newton never travelled beyond the triangle of Woolsthorpe, Cambridge and London; an assistant reported seeing him laugh once. He was a pack rat whose papers, sold at Sotheby's in 1936, revealed (to buyers like Keynes) a man obsessed with alchemy and theology as much as physics. His biographer Richard Westfall, after 20 years, found him "not finally reducible."
Born premature on Christmas Day 1642 (small enough to fit "into a quart pot"), fatherless, he was left with his grandmother when his mother remarried; he later confessed to threatening to burn his stepfather "and the house over them." At school in Grantham he built a mouse-powered windmill and a lantern kite. Taken home at 17 to farm, he failed, and went up to Trinity College, Cambridge, in 1661 as a subsizar, serving richer students though his mother had over £700 a year.
Newton worked 18-hour days and kept his results private. Mlodinow argues this isolation explains why, by 1684, at 41, he had brilliant notes and almost nothing finished: ideas not tested in a public forum stay half-formed. "Had Newton died in 1684," Westfall wrote, we would know him as a brilliant but minor figure. The same pattern appears whenever a team keeps a model or design out of review.
The plague closed Cambridge from 1665 to 1667. The legend of a single annus mirabilis and a falling apple is, Mlodinow says, destructive: Newton's insights came over 1664 to 1666, and he remained confused about force for 20 years. In his Waste Book he developed calculus (the slope of a curve as a limit, giving instantaneous speed) and over 100 axioms on motion. He won a Trinity fellowship at 24 and the Lucasian chair from Isaac Barrow (later Hawking's chair), and with prisms showed that white light is a mixture of colours: a second prism does not change coloured light. A fight with Robert Hooke over his optics made him withdraw for a decade.
Newton's famous apocalyptic calculation (Yahuda manuscript 7.3g, Hebrew University) says the prophetic period could end no earlier than 2060 ("It may end later, but I see no reason for its ending sooner"), and he explicitly disliked fixing dates. It is a lower bound, not a prediction that the world will end in 2060.
Newton spent perhaps 30 years on alchemy, with his own laboratory and library, copying recipes about a "green lion" and seeking the forces that bind matter. He wrote far more on theology than on physics, privately rejected the Trinity while a fellow of Trinity College, and considered his biblical work his most important. Mlodinow does not mock this: alchemy's idea of hidden active principles may have made Newton readier than Descartes's followers to accept a force acting at a distance. Keynes, who bought many of the papers, called him "the last of the magicians."
At a 1958 Columbia talk, Pauli presented a theory and Bohr remarked that it was not crazy enough to be correct. Mlodinow's point is that innovators produce more wrong ideas than right ones; the skill is in testing and discarding. Newton's Waste Book is full of errors that he fixed only slowly.
The optics episode shows both Newton's genius and his temper. He pushed a bodkin (a blunt needle) between his eye and the bone to see how pressure changed his vision, and stared at the sun until he had to spend days in a dark room. His prism experiments were decisive: a prism spreads white light into a spectrum, and a second prism does not change the colour of any single band, so prisms separate colours rather than create them, and white light is a mixture. His reflecting telescope won him election to the Royal Society, but when Hooke criticised his paper on colour, Newton reacted with fury and withdrew from public science for about a decade.
| Year | Event |
|---|---|
| 1642 | Born at Woolsthorpe, Christmas Day (Julian calendar) |
| 1661 | Enters Trinity College, Cambridge |
| 1665 to 1667 | Plague years at home: calculus, optics, early ideas on gravity |
| 1669 | Lucasian Professor, succeeding Isaac Barrow |
| 1672 | Reflecting telescope; paper on light and colours; quarrel with Hooke |
| 1684 | Halley's visit; De Motu |
| 1687 | Principia published |
| 1696, 1700 | Warden, then Master of the Mint |
| 1703 | President of the Royal Society |
| 1727 | Dies aged 84; buried in Westminster Abbey |
Halley's visit and the Principia
In January 1684 Halley, Hooke and Christopher Wren discussed whether an inverse-square force would produce Kepler's elliptical orbits. Hooke claimed a proof but would not show it. In August Halley asked Newton, who said he had calculated it long ago and would send it. In November came the nine-page De Motu, and then 18 months of intense work that produced the Principia (1687). Hooke's 1679 letter (an orbit as inertia along the tangent plus attraction toward the centre) helped; Newton's genius was to show that the same law governs the apple and the Moon.
- Inertia: a body continues in uniform motion unless acted on by a force.
- Force changes motion in proportion to the force (the form F = ma came about a century later).
- Action and reaction are equal and opposite (equivalent to conservation of momentum).
Plus universal gravitation: every particle attracts every other with a force proportional to their masses and inversely proportional to the square of the distance.
The inverse-square law itself has a simple geometric logic that Halley and Wren understood. A force spreading out from the Sun in all directions is shared over the surface of an imaginary sphere, whose area grows as the square of its radius; double the distance and the force is spread over four times the area, so it is a quarter as strong. The hard part was proving that such a force produces ellipses and Kepler's other laws: that a planet sweeps equal areas in equal times, and that the square of its orbital period is proportional to the cube of its average distance. Newton showed how a jagged path of tiny pushes toward the Sun becomes, in the limit, a smooth orbit.
The test was numerical. Newton predicted a body in Paris falls 15 feet and 1/8 inch in the first second and matched it to about one part in 3,000; pendulums of gold, silver, lead, glass, sand, salt, water, wood and wheat swung alike. The Principia explained tides, lunar irregularities and the precession of the equinoxes.
| Object | Terminal speed |
|---|---|
| Raindrop | About 15 mph |
| Ping-pong ball | 20 mph |
| Golf ball | 90 mph |
| Human, spread-eagled / tucked | 125 / 200 mph |
| Bowling ball | 350 mph |
| Felix Baumgartner (2012, thin air) | 843.6 mph |
Mlodinow stresses that Newton's laws describe an idealised world. Real falling bodies reach a terminal velocity when air resistance balances gravity, which is why a raindrop does not hit at the speed of a bullet.
- Alan Eustace jumped from 135,890 feet on 24 October 2014, beating Baumgartner's altitude record; but using a drogue he reached 822 mph, so Baumgartner's 843.6 mph freefall speed record still stands.
- In December 2016 a first edition of the Principia sold at Christie's for $3,719,500, then an auction record for a printed science book.
The laws become vivid in the book's examples. A Ferrari that goes from 0 to 60 in under four seconds would, if it could keep that acceleration for a year in empty space, pass half the speed of light. Pushing a 750,000-pound jumbo jet with the force you use on a 75-pound wagon would get it moving, slowly, because in Newton's world any net force produces acceleration; it is friction that hides this on Earth.
The Royal Society agreed to publish only if Halley paid, since its History of Fishes had flopped; Halley was reimbursed in unsold copies. Newton answered critics of "occult" action at a distance with "I frame no hypotheses"; Einstein's relativity later removed instantaneous action. Fame followed: MP in 1689, Warden then Master of the Mint (earning on average about £1,650 a year, some 75 times a craftsman's wage), President of the Royal Society from 1703 after Hooke died. He died on 20 March 1727 (1726 in the old calendar), aged 84, with high levels of mercury in his hair, and was buried in Westminster Abbey.
Priority fights marked the rest of his life. Leibniz had developed calculus independently, with better notation; the Royal Society's report on the dispute, which declared for Newton, was written largely by Newton himself. Hooke claimed the inverse-square law. Mlodinow notes that the lives of Galileo and Newton together span more than 160 years, and that Newton's dream, explaining all of nature by forces between particles, was realised only 250 years later, and then only with quantum laws.
Newton's end was painful. Analysis of his hair centuries later found lead, arsenic and antimony at about four times normal levels and mercury at fifteen times, as one might expect of an alchemist, but his deathbed diagnosis was a bladder stone. His fate contrasts sharply with Galileo's: instead of an obscure grave, Newton lay in state in Westminster Abbey beneath a monument showing boys holding instruments that represent his discoveries.
- Science needed institutions (universities, printing, academies) as much as genius.
- The key move was quantification: Oresme's graph, Galileo's timed balls, Newton's 1-in-3,000 match.
- Idealisation (ignore friction to see the law) is a skill, not a cheat.
- Isolation stalls even geniuses; Halley's question unlocked the Principia.
- Several famous stories (the apple, the annus mirabilis, 2060) are simplifications.
Halley knocks; the Principia answers
When Galileo published the Two New Sciences, he had brought culture only to the brink. Isaac Newton (1642 to 1727) walked through. The chapter refuses the marble-statue myth. Newton was a difficult genius: secretive, often isolated, deep into alchemy and apocalyptic chronology as well as optics and gravity. Isolation invents; it also incubates error. A closer look: his private end-of-world reckonings (sometimes popularised as "2060") are theology, not physics predictions.
| Year | Event |
|---|---|
| 1642 | Born, Woolsthorpe |
| 1661 | Trinity College, Cambridge |
| 1665 to 1667 | Plague years: calculus, optics, gravity |
| 1669 | Lucasian Professor |
| 1684 | Halley's visit; De Motu |
| 1687 | Principia |
| 1696 / 1700 | Mint warden, then master |
| 1727 | Dies; buried Westminster Abbey |
Edmond Halley's visit is the book's hinge. Newton had pieces; a visitor demanded the orbit problem and refused to leave without an answer. Conversation ships what isolation drafts. The three laws (inertia; F=ma in modern form; action-reaction) plus universal gravitation rebuilt the mechanical universe. Terminal velocity tables in the book (raindrop to Baumgartner) show air resistance capping the dream of endless acceleration.
Newton's notebooks overflow. The method is not "be Newton." The method is generate many conjectures and kill them in public. Alchemy that never meets a peer dies private; orbital mechanics that meets Halley becomes a civilisation's default metaphors (inertia, momentum, force of character).
Schedule outsider questions that demand a derivation or a null, not a status update. Thirty minutes, one question, written answer in a week. That is how De Motu becomes Principia in miniature.
Platform teams often have private notebooks of half-laws ("eventually consistent except when..."). A Halley visit forces the force law onto a page others can check.
| Object | Approx. terminal speed |
|---|---|
| Raindrop | 15 |
| Ping-pong ball | 20 |
| Golf ball | 90 |
| Human, spread-eagled | 125 |
| Human, tucked | 200 |
| Bowling ball | 350 |
| Baumgartner, 2012, thin air | 843.6 |
Isolation without Halley is a risk in remote-first orgs. Link back to Chapter 1's acceptance theme: the Principia was not instantly everyone's physics.
Newtonian psyche and the goat under a sheet
After Newton, purpose-driven nature yields to a Pythagorean universe driven by numbers. The American founders mixed Newtonian language with theology in the Declaration's "Laws of Nature and of Nature's God." Everyday speech absorbed force, acceleration, inertia and momentum. Studying Newton is studying our own metaphors.
Mlodinow's high-school confusion (why were these obvious laws hard?) is answered by his father's goat story: teenagers in Poland draped a sheet over a goat and raced it through the house as a ghost. Elders' ghosts and Newton's laws both feel natural only inside a culture. Outside it, both feel absurd. That is Chapter 1's conception problem in folk costume.
When a framework feels obvious, you are least able to see its alternatives. Schedule "pre-Newton days" on hard problems: describe the system without your default metaphors, then re-introduce them deliberately.
"The model's momentum" and "the portfolio's inertia" are Newtonian gifts. They clarify and they conceal. When a credit strategy has inertia, ask what force law would change its velocity, or whether you are naming friction as destiny.
Halley's visit remains the operating ritual. Isolation drafted; conversation shipped. Alchemy and apocalypse dates remain the shadow: private notebooks without peers.
Newton’s package; one physics for apple and moon
Calculus, optics, alchemy, Halley, Principia. Universal gravitation is the audacity that apple and moon share a law. Alchemy reminds that method filters are earned, not automatic in genius.
| Claim | Content |
|---|---|
| Laws of motion | inertia, force law, action-reaction |
| Gravitation | inverse-square, universal |
| Method | geometry + fluxions; publish the system |
Woolsthorpe, Halley and the Principia machine
Galileo brought culture to the brink. Newton built the mechanical universe. He lived inside a small triangle: Woolsthorpe, Cambridge, London. He barely "played." Diaries show an alien intensity closer to Mlodinow's father trading bread for a puzzle than to a modern lab socialite.
A Grantham schoolmaster and family allies pushed him to Trinity College in June 1661. Servants celebrated his departure. At Cambridge he absorbed the science of his time in order to overturn it. During plague years he developed pieces of calculus: for the first time, speed at a particular moment made sense. Today that toolkit describes airflow, populations, markets and weather. Then it was a private language.
Aristotelian "gravity" as an inherent property of stuff yielded to external force. Hooke and Boyle's experimental craft inspired him even when he would not admit it. He ground glass, improved telescopes and calculated. He also chased alchemy and Bible timelines; one late end-of-world window sat between 2060 and 2344. Private notebooks without peers are the shadow mode of genius.
Hooke claimed an inverse-square path to orbital proofs and never delivered. Seven months later Halley visited Cambridge. Newton said he had already proved the link between the inverse-square law and Kepler's orbital properties. Isolation drafted; the visit shipped. Principia (1687) then showed one handful of principles explaining lunar irregularities, tides, sound speed and more. Aristotle's qualitative empire fell.
Later Newton ruled the Royal Society with an iron fist, ejected "levity," and fought credit wars. The method outlived the personality.
Can five named principles, stated mathematically or operationally, generate more predictions than slides? If not, you have anecdotes, not a mechanical universe.
Halley did not invent the inverse-square idea. He asked the right question, visited the right hermit and paid for publication. Fund Halley roles: people whose KPI is "proofs shipped," not "ideas owned."
Christmas birth, wrong turns and space inertia
Newton arrived on 25 December 1642, premature, father already dead. His mother did not treasure him; popularity failed even at home. The drive to know can grow from pain as well as from bread-for-puzzle joy. After the plague annus mirabilis he also chased ideas later scholars called crazy. Trinity's 1667 election was one of those discrete forks that reshape a life.
Wren offered Hooke a prize for proving the inverse-square link to Kepler; Hooke refused, claiming he would wait so others would appreciate how hard it was. That is credit theatre. Halley's later visit still had to extract Newton's proof into the open.
Newton's first law reframes everyday experience: left alone, motion stays uniform; curves and speed changes mean force. A Ferrari fights friction to hold sixty; a spacecraft coasts. Spaceflight is Newton's intuition made vehicle. Once the mechanical universe feels natural, remember the goat under the sheet: culture made it obvious.
If someone claims a proof but will not show it "so others struggle," treat it as unproven. Hooke's refusal is the template.
Halley knocked; Principia followed. Newton had the pieces; a visitor demanded the orbit. Isolation invents; conversation ships.
Schedule a 30-minute "Halley visit" on a stalled project: one outsider, one question ("what force law fits these data?"), one written answer due in a week. No status update; only a derivation or a null.
Newton's alchemy and apocalypse dates show genius without peer review becomes private mythology. Many ideas, mostly wrong, is healthy only if wrong ones die in public.
Bridge the two chapters as one workshop: first invent a community that can lose to a measurement, then invent a single law that treats the apple and the moon as the same kind of falling. Without the first, Newton has no public to convince; without the second, Galileo's ramps stay local craft.
New way. Universities, Merton, Oresme, print, Galileo: lose arguments to nature. Mechanical universe. Newton’s mixed life; Halley; Principia; one inverse-square story upstairs and downstairs; alchemy as filter warning.
| Before | After |
|---|---|
| Separate heaven/earth stories | One gravity law |
| Qualitative motion | Geometric / fluxion proofs |
| Authority of texts | System + prediction |
Part II-a close. Measure, diagram, publish, lose to nature, then unify apple and moon. If your team cannot point to a result that forced a strategy change in the last year, you are still in the book-over-bench era with better coffee.
Chapter 8. What Things Are Made Of
As a teenager Mlodinow ran a basement chemistry lab in Chicago, and learned that errors in physics stay on paper while errors in chemistry leave scars. His father drew the same contrast: in the camp, a "physicist" was a mathematician, but a "chemist" was a saboteur. In Częstochowa he joined an underground cell that loosened railway bolts with screwdrivers; when it was betrayed, he and one other man survived by lying under a freight train.
The dangers were real for the pioneers too. The Swedish pharmacist Carl Scheele survived being the first to produce chlorine and somehow accurately described the taste of hydrogen cyanide without dying, but in 1786 he succumbed at 43 to an illness suspiciously like mercury poisoning. Early chemists tasted, sniffed and handled what they made, because they had no other instruments.
Chemistry, the chapter argues, grew from trades and magic, not from a single theory. It had no equivalent of Newton's laws, because its subject is messy: burning alone gives off more than a hundred gases, and the ancients failed to recognise that seven familiar metals (mercury, copper, iron, lead, tin, gold and silver) were elements. Progress needed tools: balances, thermometers, calorimetry, acid tests and ways to capture gases.
From embalmers to alchemists
The first chemical trade was embalming, with secret recipes of natron, resins and myrrh, which bred a culture of sorcery. In Alexandria (founded 331 B.C.), Ptolemy II's Museum paid over a hundred scholars stipends, an early research institute beside the great Library. There Greek theory met Egyptian craft. Aristotle had described elements by qualities (water wet and cold, air wet and hot), which suggested that one substance might be turned into another by changing its qualities. Alchemy, born about 200 B.C., sought gold and the elixir of life.
Paracelsus (1493 to 1541), born Theophrastus von Hohenheim, burned Galen's books, lectured in an apron in Swiss German and insisted that diseases come from outside agents treatable with specific chemical drugs; he popularised laudanum. He was also a drunk who claimed his backside had more talent than other doctors.
Robert Boyle (1627 to 1691), fourteenth child of the Earl of Cork, read Galileo's Dialogue in Florence at 15, the year Galileo died. His Sceptical Chymist (1661) argued that the products of burning wood are not elements. With his assistant Hooke, who built his air pump, he showed that animals die in sealed vessels and flames go out in a vacuum, so air is not elemental. He helped found the Royal Society (chartered 1662), motto Nullius in verba: take nobody's word for it.
Between the alchemists and Lavoisier came a technical change that is easy to miss: the pneumatic trough, which let chemists collect gases over water or mercury and so study "airs" as separate substances. Until gases could be captured and weighed, chemistry could not balance its books, since burning and fermentation seemed to make matter vanish.
Oxygen: Priestley and Lavoisier
Joseph Priestley (1733 to 1804), a Dissenting minister whose house was burned by a mob in 1791 for his support of the American and French revolutions, studied the gas over a brewery vat (carbon dioxide; he invented soda water) and in 1774 heated red mercury calx to release a gas in which a candle blazed and a mouse lived an hour instead of fifteen minutes. He breathed it and found that mint and spinach restored "spent" air (photosynthesis). But he called it dephlogisticated air, fitting it into the old theory that burning releases a substance, phlogiston.
Priestley's patron was William Petty, Earl of Shelburne, who in 1773 hired him as librarian and tutor but also built him a laboratory and left him free time. It was there, experimenting with mercury calx (mercury oxide), that he made his discovery. Priestley belonged to the Enlightenment (roughly 1685 to 1815), whose motto Kant summed up as Sapere aude, "dare to know," and to the dawn of the Industrial Revolution, when patrons and manufacturers saw profit in chemistry.
Antoine Lavoisier (1743 to 1794), rich from tax farming, met Priestley at a Paris dinner in October 1774. He did what Priestley would not: he weighed everything. Mercury heated in air gains exactly the weight of gas it absorbs; heated further, it loses it. He named the gas oxygen, established conservation of mass, devised modern chemical names and in 1789 published the first modern textbook, listing 33 elements, most of them correct.
| Priestley | Lavoisier | |
|---|---|---|
| Strength | Brilliant experimenter, new gases | Quantitative method, systematic theory |
| Interpretation | Phlogiston theory | Oxidation; conservation of mass |
| Style | "Experiments hardly interrupted by any reasoning" (Lavoisier's jibe) | A balance and a framework |
| Fate | Emigrated to America, 1794 | Guillotined, 8 May 1794 |
Lavoisier's tax wall around Paris made him hated; he was arrested in 1793 and executed. "The Republic has no need of scientists" was reportedly said at his trial, though the line is probably apocryphal. A 1900 statue of him in Paris was modelled on the wrong bust (Condorcet's head) and was melted down during the Nazi occupation.
Carl Wilhelm Scheele in Sweden isolated oxygen around 1771 to 1772, before Priestley, but his book appeared only in 1777. The book mentions Scheele (who also made chlorine and died at 43) but gives the discovery story to Priestley and Lavoisier. Historians also stress the role of Marie-Anne Paulze Lavoisier, who translated, illustrated and recorded his experiments.
Mlodinow uses a relay-race image for this history: no runner covers the whole course. Boyle showed air was not an element, Priestley isolated gases, Lavoisier weighed them and built the system, and Dalton, two decades later, explained the weights. Each depended on new instruments, from Boyle's pump to the pneumatic trough that let chemists collect gases over water.
Mlodinow's own analogy for Lavoisier's insight is an upside-down radio: an old radio he had taken apart and reassembled worked, but only when he understood every part did it make sense. Lavoisier's balance sheet of reactions was that understanding for chemistry.
| Item | Figure |
|---|---|
| Elements in his 1789 list | 33, of which all but about 10 are elements as understood today (the rest included light and "caloric") |
| Year of the Elementary Treatise on Chemistry | 1789, often called the first modern chemistry textbook |
| Age at execution | 50 (8 May 1794) |
| His tax wall around Paris | Cost the equivalent of hundreds of millions of dollars today; its tollgates were among the first targets of rioters in 1789 |
Dalton's atoms
Lavoisier had dismissed atoms as unmeasurable. John Dalton (1766 to 1844), a Quaker teacher from a weaver's family, made them measurable. How small are atoms? Imagine filling all the world's oceans with marbles, then shrinking each marble to the size of an atom: they would fit in less than a teaspoon. That such things could ever be studied seemed hopeless, and Dalton's achievement was to find an indirect route: relative weights in reactions. A teacher who rose at five for tea and kept to a strict routine, he came to atoms through meteorology, the solubility of gases and Newton's corpuscles.
From copper oxides (one with twice as much copper per gram of oxygen as the other) he inferred that elements combine in simple whole-number ratios of atoms, set hydrogen's weight to 1, and published relative atomic weights. He assumed the simplest formula (water as HO), so his oxygen weight was half the true value, but the method survived. More than 40,000 people attended his funeral.
The book dates A New System of Chemical Philosophy to 1810. Part 1 appeared in 1808 (part 2 followed in 1810), and Dalton had presented atomic weights as early as 1803.
Mlodinow pauses here on grit, which psychologists define as "the disposition to pursue long-term goals with sustained interest and effort over time" and link to success in everything from marriage to the military. Dalton, Hawking and Mendeleev all had it: discovery, he says, "can usually be won only through magnificent perseverance."
Mendeleev's table
Dmitri Mendeleev (1834 to 1907), youngest of a large Siberian family, cut his hair once a year and threw tantrums. His mother travelled about 1,400 miles with him at 15 to find a school, and died urging him to "insist on work and not on words." Writing a chemistry textbook forced him to organise the 63 known elements. Arranging cards by atomic weight in a game he called "chemical patience," he saw properties repeat periodically. He published in 1869, corrected wrong atomic weights and left gaps.
The breakthrough was a mix of work and play. Mendeleev had been appointed to a chair at 32 and needed a textbook; ordering the material forced him to look for an organising principle. According to the familiar account, he spent a Saturday arranging cards of element names and weights, then woke from a doze with the pattern. Mlodinow stresses the boldness, not the dream: Mendeleev trusted the pattern over the data, declaring several measured atomic weights wrong.
| Property | Mendeleev's "eka-aluminium" (1871) | Gallium (found 1875) |
|---|---|---|
| Density | 5.9 g/cm³ | First measured 4.7, then 5.9 after purification |
| Discoverer | Predicted | Lecoq de Boisbaudran |
He also predicted scandium and germanium, and the element that became technetium, made in 1937 in a cyclotron.
The card game matters. Mendeleev and his friends played patience, laying out cards in a set pattern; the arrangement looked to him like the table of twelve elements he had sketched on the back of an envelope. He wrote the names and atomic weights of all the known elements on cards and played patience with them. Where the pattern and the measurements disagreed, he trusted the pattern: he corrected the accepted weights of several elements and left gaps for elements no one had found.
Several chemists (Newlands in England, Meyer in Germany) noticed periodic patterns at about the same time. Mendeleev gets the credit because he made risky predictions: specific densities, weights and properties for undiscovered elements. When gallium, scandium and germanium turned up matching them, the table became a theory rather than a catalogue. In practice: a model earns trust by predicting something its authors could not have fitted after the fact.
Timing matters, Mlodinow adds: Luther Simjian's "Bankograph" cash machine of 1960 failed (its main users were gamblers and prostitutes), but ATMs took off a decade later. Mendeleev never won a Nobel; mendelevium (element 101) was made at Berkeley in 1955.
- The table has 118 named elements; the UN marked 2019 as the International Year of the Periodic Table (150 years after Mendeleev).
- In 2024 Berkeley Lab made two atoms of livermorium (116) with a titanium-50 beam, proving a non-calcium path toward heavier atoms.
- The search for element 120 (planned Ti-50 on Cf-249) is still in preparation as of October 2026: Oak Ridge fabricating californium targets, Berkeley upgrading cyclotron and detectors. Element 120 has not been discovered.
From phlogiston to the periodic table
As a teenager Mlodinow was torn between two scientific hungers: the dynamics of change and the question of what things are made of. Archimedes dreamed of levers; the young chemist dreamed of explosions. Aristotle had offered road maps for both and been wrong on both. Newton transformed change; chemistry needed its own heroes.
The path runs from embalmers and alchemists through Paracelsus and Robert Boyle (The Sceptical Chymist, 1661) to the oxygen wars. Scheele, Priestley and Lavoisier isolate and name the gas that supports combustion; Lavoisier kills phlogiston with the balance and a textbook (1789) listing about 33 elements. Priestley found the gas and clung to the old framework; discovery without framework change can miss itself.
| Priestley | Lavoisier | |
|---|---|---|
| Experimental skill | Superb | Superb |
| Framework | Phlogiston (adapted) | Oxidation, conservation of mass |
| Legacy lesson | Finding a fact ≠ updating a model | Measurement + new ontology |
Dalton (1808) makes atoms quantitative for chemistry. Mendeleev (1869) arranges elements, corrects weights and leaves gaps that are predictions. Eka-aluminium becomes gallium; density moves from 4.7 toward the predicted 5.9 g/cm³ after purification. Theory that predicts beats theory that only describes. Timing matters: Simjian's 1960 Bankograph failed; ATMs thrived a decade later. Mendeleev never won a Nobel; mendelevium (101) was made at Berkeley in 1955.
- The table has 118 named elements; the UN marked 2019 as the International Year of the Periodic Table.
- In 2024 Berkeley Lab made two atoms of livermorium (116) with a titanium-50 beam, proving a non-calcium path.
- The search for element 120 (Ti-50 on Cf-249) remains in preparation as of 2026: Oak Ridge fabricating californium targets, Berkeley upgrading cyclotron and detectors. Not discovered.
Draw your system's periodic table. Leave two deliberate empty cells where a component ought to exist. Write the test that would fill each gap. Empty slots beat crowded guesses.
Name one framework your team still uses after the data that killed it arrived. Schedule a Lavoisier meeting: balances out, ontology on the table, one postulate to retire.
Link to Chapter 9: once matter has a table, life can be asked what it is made of.
Explosions, dirt and shared limelight
Chemistry was Mlodinow's first love: ammonia with iodine, perchlorate with sugar, zinc dust with nitrates, fireworks and curiosity. Archimedes wanted a long lever; the boy chemist wanted a loud reaction. The adult lesson is sharper. Chemistry is dirtier and more complicated than celestial mechanics. It needed apparatus Newton did not have. So it produced shared limelight (Boyle, Scheele, Priestley, Lavoisier, Dalton, Mendeleev) rather than one Principia.
Newton's alchemy was not only superstition; it was a search lacking the right tools. Domain transfer fails when the tooling stack is missing. Physicists who "just apply" to biology without new instruments repeat him.
If your domain is chemistry-like (many substances, context-sensitive reactions), stop waiting for a lone Newton. Build Lavoisier's balance: shared measurement standards, open notebooks and a culture that can retire phlogiston.
New hires who can only recite architecture diagrams but cannot "mix" (run a canary, read a ledger, break a staging invariant safely) are reading alchemy texts. Give them supervised explosions.
Alchemy to table; balance sheets for matter
Paracelsus, Boyle, Priestley, Lavoisier, Dalton, Mendeleev. Mass balance and atomic weights outvote story.
Superheavy quests continue; element 120 remains unconfirmed through 2026.
Paracelsus to Mendeleev: dirty science, clean table
Teen Mlodinow loved loud chemistry and heard physics talk of elegant laws. Chemistry needed new instruments: accurate balances, heat measures, gas handling. Newton's alchemy searched without that stack. A camp story from his father (a physicist scavenging materials, vanishing on a last trip) underlines the chapter's human cost: curiosity under coercion.
Egyptians saw gold's persistence and chased transformation. Paracelsus shocked colleagues (a pan of feces as medicine's secret) and won students. Showmanship is not method; measurement is.
Boyle, raised partly in Geneva, woke at fourteen to a thunderstorm that pressed mortality on him. His sealed-vessel work with animals and flames was cruel by modern ethics and clarifying by method: air is not an Aristotelian element you hand-wave. Later gas work (Scheele, Priestley) produced "airs" that extinguished chips or sparkled in water. Discovery without explanation left the field half-built.
Lavoisier read Newton into chemistry and rejected a subject "founded on systems." He named oxygen, reframed calcination as combination, and remade the field. Revolution politics killed him; a statue later mixed his fame with Condorcet's head, a wry image of how credit and marble wander. Ideas outlived the war scrap that melted public monuments into bullets.
Dalton guessed water as HO, so his oxygen weight relative to hydrogen came out half of today's value. He knew the uncertainty and still pushed atomic weights as a program. Mendeleev hitchhiked toward St. Petersburg's Central Pedagogical Institute, wrote a textbook out of love and necessity, and organized elements by properties. Gaps became predictions (eka-elements), later filled (Boisbaudran and others). A table that forecasts emptiness is stronger than a table that only files what you already have.
If your architecture diagram has no empty cells with dates and tests, it is a filing cabinet. Mendeleev's courage was refusing to erase the holes.
Quarterly, list three theories your team treats as oxygen that might still be phlogiston. Assign a discriminating experiment to each. Retire one.
Bombast, Sapere aude and timing
Paracelsus (Theophrastus Bombast von Hohenheim, 1493 to 1541) renamed himself "greater than Celsus," studied metallurgy and alchemy, claimed medicine, and performed provocations that alienated faculty while packing student rooms. Ego is not a method; his era still needed Boyle's sealed honesty and Lavoisier's balance.
Priestley worked at Enlightenment height. Kant's motto Sapere aude (dare to know) names the cultural permission structure chemistry required. Science does not march; it is a relay where baton-grabbers often run a new direction. Lavoisier's statue with Condorcet's head is the relay's dark joke.
Timing matters: Luther Simjian's Bankograph ATM (1960) failed when only people who avoided tellers used it; a decade later the same idea scaled. Mendeleev's gaps were on time; Simjian's machine was early. Upright thinking includes calendars, not only equations.
Run a "Simjian review" on shelved ideas: was the theory wrong, or was trust and infrastructure missing? Re-test when the culture catches up.
Gaps are predictions. Mendeleev left blank cells and named what should fill them. Eka-aluminium became gallium when density moved to match the table. Empty slots beat crowded guesses.
Draw your system's "periodic table": components on one axis, properties on the other. Leave two deliberate gaps where a component ought to exist. Write the test that would fill each gap.
Priestley found oxygen and stuck to phlogiston. Discovery without framework change can miss itself. When data fight the model, retire a postulate, not just a parameter.
Chapter 9. The Animate World
Hooke named cells in 1665 for their resemblance to monks' rooms. A human body grows from one cell through over 40 doublings to tens of trillions. Biology's problem is diversity: the book cites about 10 million eukaryotic species, about 1% classified, and over 22,000 species of ants. Our food and bodies are full of other life.
- Microbes: the book says our microbial cells far outnumber our own. A 2016 re-estimate (Sender, Fuchs and Milo) found about 38 trillion bacteria to 30 trillion human cells, roughly 1.3 to 1.
- Demodex mites: the book says half of adults are free of them. A 2014 PLOS ONE study (Thoemmes et al.) found Demodex DNA on every adult sampled (19 of 19), suggesting they are nearly universal.
- Peanut butter: the FDA's defect action level is an average of 30 or more insect fragments per 100 grams, not the book's figure per 31 grams.
The diversity is easy to underrate. A scoop of good soil holds hundreds or thousands of invertebrate species and several thousand microscopic roundworms. The skin has its own ecology: the Belly Button Biodiversity project at North Carolina State University studies the life in that dark, isolated habitat, and skin mites less than a third of a millimetre long live in the hair follicles of our faces. Without an organising principle, biology was condemned to cataloguing.
Aristotle devoted perhaps a quarter of his writing to biology, believed in spontaneous generation and placed humans at the top of a ladder. The Belgian chemist van Helmont gave a recipe for making mice from wheat and dirty underwear in 21 days.
Redi used fresh snake meat, fish and veal in wide-mouthed jars, some open, some covered with gauze or paper. His reasoning was explicit: if spontaneous generation were true, maggots should appear in all the jars. Mlodinow calls it one of biology's first truly scientific experiments. Redi was also a court physician who rubbed himself with oil to ward off disease and wrote a celebrated poem on Tuscan wines, a reminder that the pioneers mixed modern method with old habits.
Francesco Redi (1668) settled the matter for insects with one of the first controlled experiments: meat in open jars bred maggots, meat in sealed jars did not, and in gauze-covered jars flies laid eggs on the gauze. (Pasteur settled it for microbes two centuries later.)
The microscope makers
The microscope was invented around the same time as the telescope; Galileo saw "flies as big as lambs." Hooke's Micrographia (1665), with 57 illustrations (a flea, a louse, a fly's eye), kept Samuel Pepys up until 2 a.m.: "the most ingenious book that ever I read."
A Delft draper and town-hall custodian with no Latin or university training, inspired by Micrographia, Leeuwenhoek (1632 to 1723) ground tiny single lenses that magnified far more than Hooke's compound instruments; he made over 500 and kept his methods secret. From 1673 he wrote hundreds of letters to the Royal Society over 50 years describing "animalcules" in a drop of water: bacteria and protists. Sceptics demanded testimonials from notaries and a pastor; Hooke eventually reproduced the sightings. Leeuwenhoek was elected a Fellow in 1680 without ever attending, and Peter the Great visited him. Lesson: outsiders can transform a field when the establishment checks their claims rather than dismissing them.
Hooke's book became a best seller during the plague of 1665 (which killed about one Londoner in seven) and the Great Fire of 1666. It also drew ridicule from doubters who could not accept that the tiny world was real; a stage comedy mocked a virtuoso who studied the anatomy of insects but knew nothing of human affairs.
Leeuwenhoek's legacy was out of all proportion to his training. His "animalcules" were whole societies of organisms, some a thousand or even ten thousand times smaller than the smallest animal visible to the naked eye. He left the Royal Society a black-and-gilt cabinet of his best microscopes; only a handful of his instruments survive, and in 2009 one sold at auction for £312,000. His letters helped found microbiology, embryology, entomology and histology, and one twentieth-century biologist called them the "most important series of communications that a scientific society has ever received."
Darwin
Charles Darwin (1809 to 1882) was a poor student with "great curiosity about facts" (he once put a beetle in his mouth to free a hand for a third). He left medicine at Edinburgh, squeamish, studied for the Church at Cambridge, where his mentor Henslow said "What a fellow that Darwin is for asking questions," and at 22 joined HMS Beagle (1831 to 1836) as the unpaid gentleman companion of Captain FitzRoy. He returned still believing the Bible.
What the experts told him was decisive. The Galápagos mockingbirds differed from island to island; the giant tortoises did too. The finches, now famous, had not been labelled by island, so they taught him little at first. Most striking was a rhea (South American ostrich) that Darwin's party had cooked and eaten before realising it was unusual and shipping its remains home: it was a new species, with its own range overlapping that of the common rhea in an intermediate zone. Closely related species replacing each other across geography suggested descent from common ancestors.
| Year | Step |
|---|---|
| 1837 | Experts report on his specimens (Galápagos mockingbirds differ by island; a new rhea); by summer he accepts evolution |
| 1838 | Reads Malthus: populations grow geometrically, food arithmetically; natural selection follows |
| 1842, 1844 | 35-page sketch; 231-page essay, with instructions to publish if he died |
| 1844 | Anonymous Vestiges (Robert Chambers) is savaged, a warning |
| 1856 | Begins his big book, Natural Selection |
| June 1858 | Alfred Russel Wallace sends the same theory from the Malay Archipelago |
| July 1, 1858 | Joint papers read at the Linnean Society to about 30 fellows; little reaction |
| Nov 24, 1859 | On the Origin of Species; first printing of 1,250 taken by booksellers at once |
Malthus supplied the arithmetic. A squid lays about 3,000 eggs; if all survived and bred, in a handful of generations the descendants would overflow the oceans. Since that does not happen, most offspring must die before breeding, and any heritable advantage, however small, would accumulate. Darwin also learned from breeders of pigeons and dogs, who achieved dramatic changes by artificial selection, and from Adam Smith's division of labour, which suggested why species diversify to fill different niches.
Ideas of evolution existed (his grandfather Erasmus had some), but they were vague. Natural selection supplied a mechanism that was testable: variation arises by chance, more offspring are born than survive, and heritable traits that help survival spread. Like Newton's laws, it removed purpose from the explanation, which is why it shook theology more than any earlier idea except heliocentrism.
The reception of Vestiges of the Natural History of Creation (1844), published anonymously by Robert Chambers, frightened him. It was a best seller and was savaged by scientists; Adam Sedgwick called it a "foul book" in an 85-page review. Mlodinow adds anecdotes from modern seminars, where scepticism can be brutal. Scientific scepticism protects against error but can also kill innovation early, which is why Darwin gathered evidence for two decades before going public.
Darwin's reason for marrying was characteristically empirical: he wrote a list of pros and cons. Emma, devout, worried that his doubts would separate them in the afterlife; Darwin said he "cryed over" her letter. It was the death of their second child, Annie, at ten in 1851, that as much as his work finally destroyed his faith. His illness, which began around 30, included vomiting, palpitations and exhaustion; he tried cold-water cures and electric belts. Hypotheses range from Chagas disease to anxiety; no diagnosis is settled.
| Darwin | Wallace | |
|---|---|---|
| Background | Wealthy gentleman naturalist, Beagle voyage | Self-taught collector who earned his living selling specimens |
| Field experience | South America, Galápagos (1831 to 1836) | Amazon, Malay Archipelago |
| Trigger for the theory | Malthus, 1838 | Malthus, 1858 (during a fever) |
| Evidence base | Twenty years of data, breeding, experiments | A brilliant short paper |
| Outcome | Origin, 1859; the theory bears his name | Gracious; later called his book on the subject Darwinism |
The famous finch "eureka" is apocryphal: Darwin's finches were unlabelled and first misidentified. The death of his daughter Annie at ten in 1851 destroyed his faith; a chronic illness plagued him from 30. He married his devout cousin Emma Wedgwood and worked from Down House in Kent, sowing seeds in brine and studying barnacles for eight years (a 684-page monograph). When Wallace's letter arrived he wrote to Lyell that "all my originality... will be smashed"; Lyell and Hooker arranged joint publication, and Wallace was gracious. Within a decade the scientific debate was largely over. Mendel's genetics, unknown to Darwin, and the molecular tools of quantum physics would later supply the mechanism of inheritance. Darwin died on 19 April 1882 and is buried near Newton.
The Linnean Society reading of 1 July 1858 was an anticlimax. The Darwin and Wallace papers were followed by others, and the members still awake after the first five were rewarded with a long treatise on the vegetation of Angola. Neither author attended: Wallace was in the Far East, and Darwin was burying his infant son, Charles Waring, who had died of scarlet fever. The Society's president later said the year had produced no "striking discoveries." Once the Origin appeared, honours followed: the Royal Society's Copley Medal, honorary doctorates offered by Oxford and Cambridge, Prussia's Order of Merit, and membership of the St. Petersburg and French academies.
The road to publication was not smooth. A reviewer for the publisher John Murray judged the manuscript "an imperfect and comparatively meager exposition" and suggested Darwin write about pigeons instead, since "everybody is interested in pigeons." Darwin declined. Exhausted, "weak as a child" when he finished in April 1859, he managed the reception like a campaign: Murray sent out many complimentary copies and Darwin mailed self-deprecating letters to the recipients. He also minimised theology, closing with a passage meant to reassure: "There is grandeur in this view of life... from so simple a beginning endless forms most beautiful and most wonderful have been, and are being, evolved."
In his last years he studied earthworms, suffered heart attacks and wrote to Wallace that "life has become very wearisome." He died in April 1882, aged 73, and was buried in Westminster Abbey with dukes, an earl and presidents of the Royal Society among the pallbearers. The Bishop of Carlisle said it would have been unfortunate to give currency to "the foolish notion" of a necessary conflict between knowledge of nature and belief in God.
Darwin's playbook is worth copying: gather overwhelming evidence, write for the general reader as well as specialists, recruit respected allies (Lyell, Hooker, Huxley) before publication, avoid needless provocation and answer critics by letter. Galileo, equally right, did the opposite and lost. Being correct is necessary for acceptance; it is rarely sufficient.
- The Darwin Correspondence Project completed its 30-volume edition of more than 15,000 letters in 2023.
- The first truly complete ("telomere-to-telomere") human genome sequence was published in 2022.
- Chemistry advanced when weighing replaced describing (Lavoisier), and when atoms became numbers (Dalton).
- Mendeleev's table won because it predicted: gaps that were later filled.
- Biology needed controlled experiments (Redi), instruments (Leeuwenhoek) and a mechanism (Darwin).
- Darwin waited 20 years, fearing the reception; Wallace forced his hand.
- Several of the book's vivid numbers (microbes, mites, peanut butter) need correcting.
Cells, selection and twenty years of caution
No one had guessed that living things, like crystals, are built from fundamental units. The most important product of a cell is a copy of itself. The sum of unthinking cellular activity is us: a galaxy of parts that somehow analyses itself. Biology is mostly invisible: cells are tiny, and diversity is vast (the book cites at least 22,000 ant species and roughly one to ten million ants per person on Earth).
Microscope makers opened the door; cell theory and germ theory walked through. Then Darwin. He accepts evolution from Galápagos evidence by 1837, reads Malthus in 1838, writes a sketch (1842) and an essay (1844), watches Vestiges get savaged, begins the "big book" in 1856, and is forced into the open when Wallace's letter arrives in June 1858. The Linnean Society hearing on 1 July 1858 barely registers; On the Origin of Species appears 24 November 1859 and sells out 1,250 copies. Twenty years of caution, then a rival's manuscript.
| Year | Step |
|---|---|
| 1837 | Accepts evolution |
| 1838 | Malthus → natural selection |
| 1842 / 1844 | Sketch and essay |
| 1844 | Vestiges savaged (warning) |
| 1856 | Big book begun |
| June 1858 | Wallace's letter |
| 1 July 1858 | Linnean Society |
| 24 Nov 1859 | Origin published |
Priority is a social sport; readiness is private. Darwin's delay was partly fear, partly rigor. Waiting forever for a perfect book is cowardice dressed as rigor. The upright move: write the two-page abstract as if a peer just emailed the same idea, then publish, partner or kill with a reason.
Cell biology and genomics have only tightened Darwin's tree: ancient DNA, CRISPR tools and AlphaFold's structure predictions (see Epilogue) make descent with modification more mechanically vivid, not less. The book's qualitative story still holds; the toolkit exploded.
If a competitor ships your half-finished idea, you feel Wallace's letter. Keep a public lab notebook (RFC, design doc) so priority and readiness do not depend on memory and panic.
Link to Part III: the atom will explain why chemistry, and thus biology, works at all.
Invisible life and the Wallace clock
Cells copy themselves; we are their sum. Biology's invisibility is double: scale and diversity. Microscope makers were the Galileo of life. Darwin supplies the law-like pattern without a legislator. The twenty-year wait is not only psychology. Vestiges showed what public failure looked like; Wallace showed what priority failure looked like. Together they created the Wallace clock: how long you may polish before a peer forces your hand.
For every private theory older than twelve months, set a date when silence becomes the risk. On that date: abstract, partner or kill. No third option called "still thinking" without a new discriminating test.
Ancient DNA, CRISPR and structure prediction make descent with modification mechanically vivid.
Microscopes to selection; life without magic generation
Redi, Hooke, Leeuwenhoek, Darwin. Catalogue plus process. Lists of species are not theories.
Name a life-claim you believe and the observation that would force a revision.
Wallace clock and invisible life
Microscopes made cells and microbes visible; Redi, Hooke and Leeuwenhoek open the aperture. Darwin supplies descent with modification after a long private build. Vestiges showed public failure modes; Wallace forced priority. Set the clock: polish has a half-life when peers exist.
Ancient DNA, CRISPR and protein structure models make the animate world's mechanisms legible far past 1859. Keep Darwin's qualitative core; update the toolkit in orange boxes, not by rewriting his argument into software marketing.
Twenty years, then Wallace's letter. Darwin had the theory; fear and polish delayed him. A rival's manuscript forced the post. Priority is a social sport; readiness is a private one.
Name one idea you have sat on for more than a year. Write a two-page abstract as if a peer just emailed the same idea. Decide within seven days: publish, partner, or kill with a reason.
Waiting for a perfect book becomes cowardice dressed as rigor. Vestiges was savaged; Darwin learned caution. Caution plus endless delay is how good ideas die unpublished.
Made of. Boyle to Mendeleev; mass balance; periodic gaps; element 120 still open. Animate. Redi to Darwin; catalogue plus process.
Write a Lavoisier ledger for one work process: what is conserved, what transforms, what measurement closes the books.
Part II-b close. Balance sheets for matter and process stories for life are the same upright demand: stop narrating substances and species as static essences. Ask what is conserved, what varies, what prediction a gap makes next.
Part III. Beyond the Human Senses
"It's the best possible time to be alive, when almost everything you thought you knew is wrong." Tom Stoppard, Arcadia (1993), the epigraph to Part III
Part III compresses the story into three decades. Mlodinow ranks the atom and its quantum laws as an innovation equal to the stone tool: every technological revolution since (television, fibre optics, phones, MRI, lasers, computers) has been a way of turning the atom into a tool. The price was that physics left human experience behind.
Chapter 10. The Limits of Human Experience
Atoms could not be seen until the late twentieth century; as late as 1900 the chemist Wilhelm Ostwald called them "hypothetical conjectures." Today "seeing" is statistical. The Higgs boson was inferred from over 300 trillion proton collisions in the 17-mile ring of the Large Hadron Collider, analysed at about 200 computing centres in three dozen countries. Mlodinow draws a parallel with art of the same era (Cézanne, Picasso, Kandinsky, Stravinsky, Joyce): around 1900, culture and physics both stopped depicting the world as it appears.
The artists' struggle mirrored the physicists'. In 1910 John Dewey wrote that critical thinking often involves "the willingness to endure a condition of mental unrest and disturbance." Mlodinow applies the phrase to creative work in both art and science: none of the pioneers of either had it easy.
François Englert and Peter Higgs shared the 2013 Nobel Prize in Physics for predicting the particle. Higgs died in Edinburgh on 8 April 2024, aged 94. The LHC continues to measure the boson's properties, which so far match the Standard Model.
Max Planck, reluctant revolutionary
In 1875 the Munich physicist Philipp von Jolly told young Max Planck (1858 to 1947) that physics was "just about complete." Planck, from a line of pastors and jurists, was "disinclined to questionable adventures." His 1879 dissertation on thermodynamics was ignored: Kirchhoff thought it wrong, Helmholtz and Clausius would not read it. He worked five years as an unpaid lecturer living with his parents.
Mlodinow dwells on those lean years because people are surprised by them: we expect artists to suffer for their art, but not scientists. His advice from Munich ranked with the builder of the Titanic calling the ship "as nearly perfect as human brains can make her": physics around 1900, like the ship, looked unassailable and was about to sink.
Every field attracts strange proposals, and most are wrong. Mlodinow's test is the one Planck faced: does the strange idea fit data that the sensible ideas cannot? Planck's formula matched every frequency Rubens checked; the "sensible" classical formula predicted infinite energy. Strangeness is not evidence for an idea, but agreement with data despite strangeness is.
The problem that made him famous was blackbody radiation (a term Kirchhoff coined in 1860): the colour spectrum of light from a hot object. Maxwell's theory of electromagnetism, the great unification of the 1860s, predicted that a hot body should radiate infinite energy at high frequencies (later called the ultraviolet catastrophe): an oven or a light bulb would be lethal.
Planck stitched together formulas that worked at low and high frequencies and announced the result at the Berlin Physical Society on 19 October 1900. Heinrich Rubens checked it against data overnight: agreement was excellent at every frequency. To explain it, Planck borrowed the statistical methods of his opponent Ludwig Boltzmann, who split energy into small portions as a calculating trick and then let their size go to zero. Planck kept the portions finite: energy comes in quanta, E = h × frequency. He announced this on 14 December 1900, the birth of quantum theory, and won the 1918 Nobel. He thought of it as a formal device; even he did not believe in atoms at first.
Mlodinow likes to book restaurant tables under the name "Max Planck." Once a maître d' told him quantum theory was invented in 1960 by the Manhattan Project. The anecdote leads to a question he returns to: are such laws invented or discovered? His answer is both. The mathematics is a human creation, but nature decides whether it fits.
Reaction was muted. Sir James Jeans, who had worked on the problem without deriving the full formula, conceded that Planck's law agreed with experiment while his own, obtained by setting h = 0, could not, yet still argued that h = 0 was the only reasonable value. Mlodinow pairs this with Robert Frost's 1914 lines: "Why abandon a belief / Merely because it ceases to be true." For five years almost no one followed Planck up.
"A new scientific truth does not triumph by convincing its opponents... but rather because its opponents eventually die" (often shortened to "science advances one funeral at a time"). Mlodinow links it to incumbents who could not accept change: video-rental chains, record labels and bookstores against Netflix, YouTube and Amazon. Planck himself was an exception: open to his opponents' methods, he converted.
Einstein's 1905
Albert Einstein (1879 to 1955) was told by a Munich teacher he would never amount to anything, left school at 15, failed the entrance exam to Zurich's ETH, relied on his friend Marcel Grossmann's lecture notes and graduated with references so poor that his father's 1901 letter asking Ostwald for a job went unanswered. (Ostwald later became the first to nominate him for a Nobel.) In 1902 Grossmann's father got him a job at the Bern patent office, where he was denied promotion in 1904. In 1905 he published three papers that changed physics.
| Paper | Claim | Consequence |
|---|---|---|
| Light quanta ("On a Heuristic Viewpoint") | Light itself comes in packets; explains the photoelectric effect | Einstein called it "very revolutionary"; earned the 1921 Nobel |
| Brownian motion | Pollen jitters because molecules bombard it | Convinced sceptics, including Ostwald, that atoms are real |
| Special relativity | Light speed is the same for all observers; no universal clock | Einstein called it "no revolution at all" |
Einstein also benefited from his obscurity. He found patent examining stimulating and said it left him "eight hours of idleness" each day for physics; he also stole time at work, shoving calculations into a drawer when anyone came near. Mlodinow's father offered a homelier lesson about why new ideas are resisted: in prewar Poland he and friends found a deer killed on the road and ate it, seeing nothing wrong, while Americans find roadkill disgusting because they were raised to. Many beliefs are habits of upbringing, and physicists are no exception.
| Item | Figure |
|---|---|
| Einstein's age | 26 |
| Papers that changed physics | 3 (plus his doctoral thesis on molecular sizes) |
| Year "photon" was coined | 1926 |
| Years until the light-quantum paper won a Nobel | 16 (the 1921 prize, awarded in 1922) |
| Photons per billionth of a second from a 100-watt bulb | About a billion |
In the photoelectric effect, brighter light releases more electrons but not faster ones; only higher frequency raises their energy. That makes sense only if light arrives in packets (named photons in 1926). A 100-watt bulb emits about a billion photons every billionth of a second. Even in 1913 Planck and colleagues, recommending Einstein to the Prussian Academy, excused his having "missed the target" with light quanta. Robert Millikan, trying to disprove the idea, confirmed it. Einstein himself wrote in 1951: "What are light quanta?" He took up the quantum at 25 and by 50 opposed what it had become.
Special relativity, the least revolutionary paper in Einstein's own view, still overturned common sense. Maxwell's equations imply that light always travels at the same speed. Einstein took that seriously: a vendor on a moving train and an observer on the platform must then disagree about distances and the timing of events. There are no universal clocks or metre sticks; each observer carries a personal measure of space and time. Newton's laws survive as an excellent approximation at everyday speeds. The paper inspired the young Heisenberg (who read Einstein's popular book on relativity) and gave Bohr the courage to break with classical physics.
The book lists James Peebles among great physicists the Nobel committee passed over. He won the 2019 Nobel Prize in Physics for theoretical discoveries in physical cosmology, sharing it with Michel Mayor and Didier Queloz (first exoplanet around a Sun-like star). The book's example of Lise Meitner stands: she co-discovered fission, but Otto Hahn alone received the 1944 chemistry prize; element 109, meitnerium, was named for her in 1997.
Anomaly first: the door to the quantum
Two million years after stone tools, humans learn to turn the atom into a tool. Ovens, clocks, fluorescent bulbs and nuclear technologies all depend on quantum rules alien to everyday experience. Chapter 10 is the hinge into Part III: classical physics fails in a corner, and the upright response is to take the corner seriously.
The isolated failure was the blackbody spectrum. Classical theory blew up at high frequencies (the ultraviolet catastrophe). Max Planck, on 19 October 1900, stitched working formulas together; Rubens checked overnight. On 14 December 1900 Planck kept energy portions finite: E = h × frequency. He treated quanta as a calculating trick; he won the 1918 Nobel anyway. The chapter's lesson is methodological: persistent anomalies are messages, not noise to smooth away.
- Check the instrument.
- Check the operator.
- Only then suspect the model. Escalate to new ontology only after the anomaly survives better tools and hostile review. Worshipping anomalies breeds crackpot theories; ignoring them freezes science in Aristotle's chair.
Mlodinow notes that people speak of technology, computer, information and nuclear revolutions, but they boil down to understanding and harnessing the atom. Fluorescent light is electrons making quantum jumps. The great revolution dates to the early twentieth century, yet its first crack was not immediately seen as a red flag. That lag is Chapter 1's acceptance theme in lab clothing.
List the top three "should not happen" events in your systems this quarter. Assign each: measurement error, operator error or model error. One owner per anomaly. No smoothing until classified.
| Classical expectation | Quantum pressure | |
|---|---|---|
| Energy | Continuous | Packets (quanta) |
| Light | Wave (mostly) | Also particle-like (next chapters) |
| Attitude to outliers | Smooth or ignore | Interrogate |
| Hero type | Completer of Newton | Reluctant revolutionary (Planck) |
Planck did not set out to overthrow physics. He patched a curve and kept the patch finite. Many upright advances begin as reluctant fixes. Celebrate the patch that forces a new vocabulary.
Link to AI: large models are theories of observables until interpretability catches up; anomalies in eval sets are your blackbody curves.
Atoms as tools and the twenty-year struggle
Stone tools were the first harnessing of nature; the atom is the equal epiphany. Fluorescent light, ovens, clocks and nuclear tech are quantum engineering. Nineteenth-century experiments saw only averages of immense numbers of atoms. Were unobservables real? Many said no. Verifiability discarded untestable ancient talk and also, for a time, threatened atoms themselves.
Blackbody radiation looked like confusion about applying Newton, not a red flag. The quantum struggle then ran about twenty years. Causality and chance arrive late in that drama; Chapter 12 finishes them. Chapter 10's job is to make you respect the anomaly and the patience.
Atoms passed from philosophy to science when predictions about averages and, later, single-particle effects landed. Your hidden state variables deserve the same bar: predict something hostile parties can measure.
Quantum door; 1905 as prosthetic year
Planck; Einstein 1905 cluster. Human intuition becomes a province. Invent the prosthetic sense nature requires.
SENSE → PROSTHETIC → EQUATION.
Blackbody, 1905 and the unobservable bargain
Nineteenth-century experiments saw averages of immense atom counts. Many denied unobservables on principle (positivism, Mach-flavored caution). Verifiability cleaned out empty talk and, for a time, threatened atoms themselves.
Blackbody radiation looked like a bookkeeping failure in applying classical ideas (the so-called ultraviolet catastrophe in later teaching), not an obvious revolution. Planck introduced energy elements to fit the curve; he was not eager to overthrow Newton. The quantum struggle then ran roughly twenty years.
Einstein's 1905 cluster (photoelectric light quanta, Brownian motion as atomic jiggling, special relativity) made unobservables earn their keep by hostile predictions. Light quanta explained Hertz-class photoelectric puzzles; Brownian motion tied to atomic sizes; relativity rebuilt space and time measurements. E equals m c squared entered the culture later as the most famous formula from that burst.
A hidden variable is allowed only if it forecasts a number an opponent can measure without sharing your metaphysics. Planck's fit, Einstein's photoelectric slope, Perrin's Brownian numbers: same bar.
Cubism and modern physics, Mlodinow notes, both stopped painting the world only as it appears around 1900. Instruments and maths became the new eyes. Later cosmology (Penzias and Wilson's microwave hiss, and today's DESI hints) continues that bargain: trust the apparatus when the senses are blind.
Latent dimensions are your atoms. Demand a measurable downstream effect (calibration curve, A/B lift, audit metric) before you treat them as real in a credit or hiring model.
Stone tools, unpaid lecturers and personal reality
Two million years ago, stone cutting tools were the first great harnessing of nature. About a hundred years before Mlodinow wrote, quantum theory became a peer epiphany in how we live (light, electronics, later computing). Planck, after unpaid lecturing and collecting student fees like a young Mendeleev, introduced quanta as a fitting device; many hoped a classical route would erase them. Boltzmann's statistical world sat nearby, still contested.
Relativity cracked Newton's single objective stage. Space and time were no longer a fixed theatre all observers share like God watching from outside. Each observer's measurements of the play can disagree in disciplined ways. That is not "anything goes"; it is a stricter bookkeeping of clocks and rods.
Ask whether your project is a stone-tool shift (new harness) or a notation tweak. Fund them differently. Quantum was a harness; another dashboard is not.
Anomaly first, theory later. Blackbody curves broke classical physics before anyone had quantum. Treat persistent outliers as messages, not noise to smooth away.
List the top three "should not happen" events in your logs this quarter. For each, ask: is this measurement error, operator error, or a model error? Assign one owner to prove which.
Worshipping anomalies breeds crackpot theories. Planck kept quanta as a calculating trick at first. Escalate only after the anomaly survives better instruments and hostile review.
Chapter 11. The Invisible World
Mlodinow's 2015 chapter title used a synonym for domain that this style guide bans; we keep the argument under Invisible World.
As a young Caltech researcher, Mlodinow watched Richard Feynman tell a visitor who had spent 12 years on a pet theory, "That's exactly what I mean about wasting your time." The lesson: choose problems you can solve, and accept that physics, unlike art, can prove you wrong.
Mlodinow had seen the problem from the other side. When a visitor from a Montreal university objected that young people should not be discouraged from theories the establishment rejects, Feynman held firm. Mlodinow's lesson is balance: be bold in the problems you choose without being so careful that you never do anything new. That, he says, is why tenure matters to science: it "makes it safe to fail."
The experimental era began with Röntgen's X-rays (1895), J. J. Thomson's electron (1897) and Ernest Rutherford's work on radioactivity (1899 to 1903), which showed that atoms transmute. Niels Bohr (1885 to 1962) wrote a dissertation criticising Thomson, went to Cambridge in 1911, found Thomson would not read it and moved to Manchester to work with Rutherford: loud, New Zealand-accented and impatient with contradiction, against Bohr's soft voice and love of argument.
The early reception of radioactivity shows the book's second theme. When Rutherford and Frederick Soddy claimed that one element turns into another, colleagues suspected them of "pulling their legs"; transmutation sounded like alchemy. Thomson won the physics Nobel in 1906 and Rutherford, to his amusement, the chemistry prize in 1908.
Transmutation itself was later achieved: in 1941 physicists turned mercury into gold, literally the alchemists' dream, by bombarding it with neutrons. In the early years, though, colleagues treated Rutherford's claim as if he had "grown a long beard and donned an alchemist's robe."
The gold foil
Rutherford's team fired alpha particles (helium nuclei) at thin gold foil. Thomson's "plum pudding" atom, a diffuse positive sphere with electrons embedded, predicted only slight deflections. Counting faint scintillations in a dark basement was tedious (Rutherford quit after two minutes; Hans Geiger was "a demon" at it). In 1909, at Rutherford's suggestion, an undergraduate, Ernest Marsden, looked for large-angle scattering. A few alpha particles bounced straight back: "as if you fired a 15-inch shell at a piece of tissue paper and it came back and hit you."
Geiger, ironically, would later make his own counting skill obsolete by inventing the Geiger counter. Rutherford's suggestion to look for large angles came, Mlodinow notes, after he had been sitting in on a probability class in the mathematics department and realised that a few particles might be scattered more than his apparatus was designed to detect. Over the following years the team counted more than a million scintillations.
| Quantity | Value |
|---|---|
| Density | A cupful of nuclear matter would weigh about 100 times as much as Mount Everest |
| Size | A minuscule dot; the atom is mostly empty space |
| Classical collapse time for an orbiting electron | Roughly a hundred-millionth of a second |
| Year Rutherford announced the nuclear atom | 1911 |
The atom had a tiny, dense, positive nucleus. But a miniature solar system is unstable in classical physics: orbiting electrons should radiate energy and spiral into the nucleus in about a hundred-millionth of a second.
Bohr's atom
Bohr's answer (published July 1913, at 27) was audacious and partly ad hoc. In hydrogen, the single electron can occupy only certain allowed orbits (stationary states) with fixed energies. It does not radiate while in one. When it jumps between orbits it emits or absorbs a photon whose frequency equals the energy difference divided by Planck's constant. A lowest orbit (the ground state) prevents collapse. In February 1913 a friend reminded him of the Balmer formula for hydrogen's spectral lines, and the model reproduced it exactly.
Bohr's priorities were clear. He married Margrethe Nørlund on 1 August 1912 but cancelled their honeymoon in Norway and spent it in a Cambridge hotel room dictating a paper to his new wife. The theory was admittedly a hodgepodge: it combined classical orbits with quantum rules and gave no mechanism for jumps.
Lord Rayleigh could not believe "Nature behaved in that way," while adding, presciently, that "men over seventy should not be hasty in expressing opinions on new theories." Arthur Eddington had already dismissed the quantum ideas of Planck and Einstein as "a German invention."
Reactions split. Sommerfeld embraced it; Einstein called it "one of the greatest discoveries"; a Göttingen physicist called it "awful nonsense, bordering on fraud"; Rutherford asked how the electron knew where to jump (in quantum mechanics, the question has no answer). Between 1913 and 1923 the model was vindicated: atomic number (protons), not atomic weight, orders the periodic table, which explained Mendeleev, and Bohr's theory helped predict hafnium (1923, named for Copenhagen).
The book gives the atomic-number story to Bohr's circle. Henry Moseley's 1913 X-ray measurements established atomic number experimentally (he was killed at Gallipoli in 1915), and hafnium was found by Dirk Coster and George de Hevesy. Bohr's model itself was wrong in detail (electrons do not orbit), a useful scaffold for quantum mechanics.
Observables only: Rutherford to Heisenberg
Just after his Ph.D., Mlodinow fished for a topic at Caltech and brushed against string theory's fashion and Feynman's caution. A visitor from Montreal defended working twelve years on an unaccepted theory; Feynman warned about wasting time without forbidding novelty. The chapter uses that scene to frame the early quantum decades: brilliant guesses, incomplete theory and the need to drop what you cannot see.
J. J. Thomson finds the electron (1897). Einstein (1905) treats light as quanta to explain the photoelectric effect and wins the 1921 Nobel for it, not for relativity. Rutherford (1911) scatters alpha particles and finds a tiny nucleus; most of the atom is empty. Bohr (1913) quantises electron orbits to save the nuclear atom from radiating itself to death. Each step discards a comforting picture.
Build theory from observables. Unobservable electron orbits were the trap. If a variable never appears in a measurement plan, cut it or add the metric. The dark side: instrument myopia. Pair the cut with Rutherford's stubborn scattering that forced a new picture despite what old instruments "could not see."
François Englert and Peter Higgs shared the 2013 Nobel for the mechanism. Higgs died in Edinburgh on 8 April 2024, aged 94. LHC measurements still match the Standard Model. The particle's discovery is Chapter 11's spirit in modern dress: predict from theory, then force nature to answer with a machine.
The invisible world is not mysticism. It is the claim that reliable knowledge can outrun the human senses if instruments and maths are honest. Cubism and modern physics, the book notes, both stopped depicting the world only as it appears around 1900.
Allow a fraction of time for unaccepted ideas; require a kill criterion and a calendar. Twelve years with no discriminating test is not courage. It is a sunk-cost shrine.
Large models map inputs to outputs without a readable mechanism. Interpretability research is a Heisenberg project: rebuild understanding from what can be probed. Do not confuse "we cannot see it yet" with "it is not real," or "the dashboard lacks a panel" with "the risk is gone."
Link back to Aristotle: common sense is not evidence. Link forward to matrices, waves and uncertainty.
String caution and the nuclear atom
Feynman's warning to young Mlodinow (do not waste years on an unaccepted theory without care) sits beside a Montreal visitor's twelve-year private project. The upright balance: protect novelty, demand kill criteria. Rutherford's gold foil forced emptiness and a nucleus; Bohr patched stability with postulates that worked and begged for a deeper why. Heisenberg later cut unobservable orbits. Einstein's photoelectric light quanta won a Nobel while relativity made him famous elsewhere.
The LHC and Higgs discovery (Englert and Higgs, 2013 Nobel; Higgs died 2024) continue Rutherford's spirit: build a machine that makes nature answer a precise question. Budget for machines (or their software analogues) not only for narratives.
Nucleus, Bohr; scattering as sight
Rutherford, Bohr. Angles and spectra replace armchair substance talk. Title avoids banned lexicon: invisible world of scales.
Bohr’s atom was a ladder. Do not worship a rung.
Nucleus, Bohr and Feynman caution
Röntgen's X-rays (1895), Thomson's electron (1897) and Rutherford's radioactivity and gold-foil work emptied the atom and packed a nucleus. Bohr quantized orbits with postulates that worked and demanded a deeper why. Heisenberg would later cut unobservable paths. Feynman's advice to young Mlodinow (do not burn years on an unaccepted pet theory without open eyes) pairs with a visitor's twelve-year private project: novelty needs kill criteria.
Mendeleev's gaps still structure the search for superheavies. Element 120 (systematic name unbinilium until discovery) remains undiscovered as of October 2026; titanium - californium campaigns continue (including planned Cf-249 target work). The table still predicts ahead of the lab.
Drop what you cannot see. Heisenberg built matrix mechanics from observables only. Unobservable orbits were the trap. If a variable never appears in a measurement plan, cut it.
Audit one model (risk, ranking, capacity). Circle every internal state that never maps to a logged metric. Delete or demote those states in the next revision, or add the metric.
"Only observables" can become instrument myopia: if your tools cannot see it, you declare it unreal. Pair Heisenberg's cut with Rutherford's stubborn scattering that forced a new picture.
Chapter 12. The Quantum Revolution
By the early 1920s the quantum was "a hopeless mess" (Max Born); Wolfgang Pauli wished he were a film comedian. Bohr opened his Copenhagen institute in 1921 while Germany slid into chaos (a kilo of bread cost 500 billion marks in 1923).
The despair had a political backdrop. Germany's Foreign Minister Walther Rathenau was assassinated in 1922, and in the 1923 hyperinflation a kilo of bread cost hundreds of billions of marks. Physicists nevertheless kept a faith Mlodinow considers essential to science: that rules exist and can be found.
Werner Heisenberg (1901 to 1976), a classics professor's son and a fierce competitor, was told by the mathematician Lindemann (who had proved pi transcendental) that reading Weyl's relativity book meant he was "completely lost to mathematics," so he went to Sommerfeld for physics. His insight, during a hay-fever retreat on the North Sea island of Heligoland in 1925, was radical: abandon unobservable orbits and build a theory only of what can be measured (spectral frequencies and intensities), arranged in arrays for each pair of states. Born recognised the arrays as matrices; with Pascual Jordan they published the "three-man paper" in November 1925. Pauli used it to derive the hydrogen spectrum.
| Heisenberg (1925) | Schrödinger (1926) | |
|---|---|---|
| Form | Matrix mechanics: arrays of observable quantities | Wave mechanics: a differential equation for a wave function |
| Picture | No picture; refuses to describe unobservables | Seemed visualisable, like classical waves |
| Reception | Einstein: an "indirect description of reality" | Einstein: "true genius"; most physicists adopted the method |
| Resolution | Schrödinger proved them mathematically equivalent (May 1926); Dirac unified the formalism, and Feynman added a third (path integrals) two decades later |
Heisenberg's path to that idea ran through failure. At Copenhagen in 1924 he watched Bohr try, and fail, to do away with photons in the Bohr-Kramers-Slater theory. Mlodinow uses the episode to make a point about heroes: Bohr was revered as having "a direct line to God," yet he was often wrong, and the young learned that even giants guess. Michael Jordan's Nike advertisement makes the same point: he missed more than 9,000 shots.
Heisenberg's interview with the mathematician Ferdinand von Lindemann went badly partly for comic reasons: Lindemann, near retirement and partly deaf, kept a poodle on his desk that barked so loudly he could hardly hear his visitor. What doomed the application was Heisenberg's mention of Weyl's book on relativity. He went instead to Arnold Sommerfeld, who told him, "It may be that you know something; it may be that you know nothing." He took his doctorate in 1923, worked under Born in Göttingen and joined Bohr in Copenhagen in 1924.
Erwin Schrödinger (1887 to 1961) produced his wave equation in early 1926 after a Christmas holiday in Arosa with a mistress, in what a colleague called a "late erotic outburst." The two founders loathed each other's versions (Heisenberg called Schrödinger's "disgusting"), yet they were the same theory. By 1960 more than 100,000 papers had applied quantum mechanics.
Position and velocity cannot both be known precisely, and the limit is nature's, not our instruments'. For a soccer ball the limit is absurdly small; for an electron confined to an atom, velocity is uncertain by about 1,000 km/h (book's figure). Quantum mechanics predicts probabilities, not certainties, ending Newtonian determinism. Einstein objected that God "does not play dice."
Einstein thought quantum mechanics incomplete, hiding deeper variables. John Bell (1928 to 1990) turned the debate into an experiment, and tests ruled out local hidden variables: "quantum uncertainty is here to stay."
- In 2015 three groups (Delft, NIST and Vienna) performed loophole-free Bell tests. The 2022 Nobel Prize in Physics went to Alain Aspect, John Clauser and Anton Zeilinger for experiments with entangled photons violating Bell inequalities.
- The UN declared 2025 the International Year of Quantum Science and Technology, marking the centenary of Heisenberg's 1925 paper.
- The 2025 Nobel Prize in Physics went to John Clarke, Michel Devoret and John Martinis for showing (1984 to 1985) macroscopic quantum tunnelling and quantised energy in a superconducting circuit, the basis of today's superconducting qubits.
- In December 2024 Google's 105-qubit Willow chip showed error rates falling as the error-correcting code grew larger ("below threshold"), a key step toward useful quantum computers.
Einstein never accepted the verdict. His late years became, in his own phrase, a "solitary little song," and in 1949 he described himself as regarded as a "petrified object." Born called his old friend's opposition a "hard blow." The founders of quantum theory thus repeated the book's pattern: the revolutionary of 1905 became the resister of 1927.
Mlodinow's father gives the uncertainty principle an unexpected human echo. During a roundup, he hid in a latrine with an underground leader he had been assigned to protect; the stench drove them out, the fugitive bolted and was never seen again, and his father joined the end of the line. Life, the son reflects, is full of randomness; quantum randomness is easy to accept by comparison.
Diaspora
Hitler became chancellor in January 1933. The book says about 2,000 leading scientists fled within five years. The fates of the founders diverge: Rutherford organised refugee aid and died in 1937; Einstein, in Pasadena when Hitler took power, never returned and became a US citizen in 1940; Born went to Edinburgh and, finally, won the 1954 Nobel; Schrödinger moved to Oxford (with wife and mistress) and then Dublin; Bohr fled Denmark in 1943 in the bomb bay of a Mosquito, passed out from lack of oxygen and advised the Manhattan Project. Of those who stayed, Jordan joined the storm troopers and was the only pioneer without a Nobel; Planck's son Erwin was executed after the July 1944 plot; Heisenberg led the German nuclear programme, was interned at Farm Hall with nine others after the war and died in Munich in 1976.
| Person | After 1933 | Died |
|---|---|---|
| Einstein | Left Germany for good; Princeton; US citizen 1940; his brain is now at the National Museum of Health and Medicine, Silver Spring | 1955 |
| Born | Barred from teaching; Cambridge, then Edinburgh; Nobel 1954; tombstone reads pq - qp = h/2π | 1970 |
| Schrödinger | Left Berlin in protest; Oxford, then Dublin; Nobel 1933 with Dirac | 1961, aged 73 |
| Pauli | Ran a refugee project from Zurich; denied Swiss citizenship, went to the US; Nobel 1945; co-founded the C. G. Jung Institute | 1958, aged 58 |
| Dirac | Lucasian Professor; declined the Manhattan Project; Florida State University | 1984, aged 82 |
| Bohr | Escaped Denmark in 1943 in a Mosquito's bomb bay; Manhattan Project adviser | 1962, aged 77 |
| Heisenberg | Stayed; led German fission research; never joined the party | 1976 |
| Planck | Stayed; met Hitler in 1933 to plead for Jewish colleagues; son Erwin executed in 1945 | 1947, aged 89 |
In 1988, on the set of MacGyver in Vancouver, Mlodinow got a call: his father, 76, had a stroke after surgery on his aorta and was in a coma. The doctor's grim prediction came true. "I hated science just then." Yet knowledge gave him strength: the father's atoms would disperse but not vanish. He promised to write this book. The memory he closes with is from the camp: his father hid in a latrine with a fugitive resistance leader while the SS counted prisoners; the count stopped at 3,000 of 3,004. The rest were shot.
| Person | Key contribution | Nobel |
|---|---|---|
| Max Planck | Energy quanta (1900) | 1918 |
| Albert Einstein | Light quanta, photoelectric effect (1905) | 1921 |
| Niels Bohr | Quantised atom (1913) | 1922 |
| Werner Heisenberg | Matrix mechanics, uncertainty (1925, 1927) | 1932 |
| Erwin Schrödinger, Paul Dirac | Wave mechanics; unified formalism | 1933 |
| Wolfgang Pauli | Exclusion principle | 1945 |
| Max Born | Probability interpretation | 1954 |
| Pascual Jordan | Co-author of matrix mechanics | None |
Heisenberg's own wartime record remains contested. After an SS newspaper attacked him as a "white Jew" for teaching relativity, he appealed to Heinrich Himmler through their mothers' acquaintance; an eight-month investigation cleared him, on condition that he not name Jewish physicists such as Einstein in public. He never joined the Nazi party. Mlodinow admits he skipped a 1973 Harvard talk by Heisenberg in protest, and later worked as a Humboldt fellow at the Max Planck Institute Heisenberg had directed.
Mlodinow closes the chapter by weighing what the quantum pioneers left. After Newton, scientists in every field wanted to "Newtonise" their subjects; quantum physics ended that dream with a truth that is "both empowering and profoundly humbling." Empowering, because we can understand and manipulate an unseen world beyond experience, the basis of computers, the Internet, satellites and phones. Humbling, because nature itself sets limits on what we can know and control. Heisenberg's theory was the culmination of long threads: more than two thousand years after the idea of the atom, two hundred after Newton's mechanics and twenty after Planck. Given the exponential growth in the number of scientists, Mlodinow expects the next hundred years to bring advances as great as the last thousand: there are, as Hamlet told Horatio, "more things in heaven and earth" than our philosophy has dreamt of.
Matrices, waves and a hard century
By the early 1920s the quantum had facts without a general theory. Physicists kept faith that nature is governed by general rules, not a hodgepodge. Those whose faith was weak dropped out; those who stayed faced both maths and politics. Heisenberg (1925) builds matrix mechanics from observables. Schrödinger (1926) offers wave mechanics. The fight is real; equivalence comes later. Uncertainty (1927) states a limit, not a complaint about instruments. Dirac and others unify the language.
| Year | Step |
|---|---|
| 1897 | Thomson, electron |
| 14 Dec 1900 | Planck, energy quanta |
| 1905 | Einstein, light quanta |
| 1911 | Rutherford, nucleus |
| 1913 | Bohr, quantised atom |
| 1925 | Heisenberg, matrix mechanics |
| 1926 | Schrödinger, wave mechanics |
| 1927 | Uncertainty principle |
Rival formalisms can describe the same physics. When two teams use different languages for one system, build a Rosetta table. If fewer than most concepts map, you do not yet share a problem. Aesthetic preference ("nature must be continuous") is a clue, not a veto.
- 2015: loophole-free Bell tests (Delft, NIST, Vienna).
- 2022 Nobel: Aspect, Clauser, Zeilinger for entanglement experiments.
- 2025: UN International Year of Quantum Science and Technology (centenary of Heisenberg 1925).
- 2025 Nobel: John Clarke, Michel Devoret and John Martinis for macroscopic quantum tunnelling and quantised energy in superconducting circuits (1984 to 1985 work), foundation of superconducting qubits.
The diaspora timeline matters. Many founders lived through 1933 to 1945 under threat; some left Germany, some stayed in compromised positions. The book refuses to separate the ideas from the century that forged and nearly destroyed their community.
Identify one dispute that persists because each side measures a different proxy. Design a loophole-light test both sides pre-accept. Ship the test before another round of narrative.
| Einstein | Bohr / Copenhagen spirit | |
|---|---|---|
| Hope | Hidden completeness | Probability as fundamental |
| Style | God does not play dice | Complementarity, observables |
| Legacy | Relativity; photoelectric | Working quantum practice |
| Lesson | Beauty can block acceptance | Working rules can outrun ontology |
The drive to know did not stop when the answers stopped looking like tables and ramps.
Faith, hyperinflation and matrices
Early quantum workers trusted that a general theory existed. Weak faith dropped out. Born and Pauli's despair makes sense: hard maths plus hard times. In 1922 Germany's foreign minister was assassinated; in 1923 the mark sank toward a trillionth of prewar value; bread cost hundreds of billions of marks. They still sought sustenance in the atom. Heisenberg, age twenty-three in 1925, competitive since childhood chair-fights with his brother, attacked the puzzles with the same fierceness. Matrix mechanics, then waves, then uncertainty: the city got two maps and a speed limit.
Faith that an answer exists differs from faith that my answer is final. The first keeps you in the game; the second blocks Chapter 1's acceptance work. Bell tests and the 2022 and 2025 quantum Nobels reward the first kind.
When your company has a "1923" (currency crash of trust, reorg hyperinflation), protect a small team whose job is still the atom: the hard unanswered technical question. Without them you only manage the crash.
Uncertainty, diaspora; formalism and memory
Heisenberg, Schrödinger, uncertainty; exile and murder; father’s death. Hold the theory and the people.
Nobel Physics 2025 (Clarke, Devoret, Martinis): macroscopic quantum tunnelling and quantisation in circuits.
Matrices, waves and the diaspora
By the mid-1920s the quantum city needed a map. Heisenberg (matrix mechanics, 1925), spurred by Sommerfeld's school and fierce competitiveness, dropped unobservable orbits. Schrödinger found waves (Arosa sickbed legend included). Born gave the probability reading. Pauli enforced exclusion. Uncertainty (1927) set a speed limit on joint knowledge of conjugate quantities. Copenhagen and Solvay meetings turned private notebooks into a public language; Einstein resisted the dice, and the argument sharpened the theory.
Then politics. Hitler's rise, Aryan physics rhetoric and the diaspora scattered the builders. Allied work and exile careers rewired where quantum craft lived. Mlodinow's family story (including camp and loss) is not ornament; it is the chapter's reminder that upright thinking is a fragile institution, not a law of nature.
Matrix and wave formulations looked like rivals; they became translations. When two teams fight over notation, build a Rosetta table before you build a reorg.
The 2025 Nobel Prize in Physics honored John Clarke, Michel Devoret and John Martinis for macroscopic quantum tunnelling and energy quantisation in electrical circuits (1984 to 1985 Josephson work). Billions of Cooper pairs behaving as one quantum system underwrites today's superconducting qubits. Chapter 12's "faith that an answer exists" now includes engineered quantum objects you can wire on a chip.
Three minds and refugee physics
Bohr's atom still assigned classical-looking position and velocity to electrons experimenters could not watch. Heisenberg cut those unobservables; Schrödinger offered waves; decades later Feynman added path integrals, a third mathematically equivalent mind. Wallace Stevens's "I was of three minds" fits the quantum city: one skyline, three maps.
Under Nazi rule, "Jewish physics" rhetoric tried to unperson creators. Rutherford at Cambridge aided academic refugees before dying in 1937 after delaying hernia surgery. Dirac held Newton's Lucasian chair and worked through the storm. Institutions that host refugees of ideas keep the upright craft alive when states turn on it.
If a dominant faction labels a method "enemy physics," document the predictions that still work. Protect a small group that can still run the discriminating tests.
Matrices and waves, same physics. Heisenberg and Schrödinger fought, then Dirac and others showed equivalence. Rival formalisms can be two maps of one city.
Take a debate where two teams use different languages for the same system (SQL vs events; control vs ML). Write a Rosetta table of ten concepts. If fewer than seven map cleanly, you do not yet share a problem.
Aesthetic preference ("nature must be continuous") blocked Einstein for decades on quantum probability. Beauty is a clue, not a veto. Let the Bell tests of your domain decide.
Keep a short “beauty veto” log: when a design feels too ugly to be right, write the measurement that would force you to accept it anyway. Run that measurement before you kill the design on aesthetics alone.
Epilogue
A monk walks up a mountain one day and down the next, along the same path. Is there a spot he passes at the same time of day on both trips? Algebra is hard; a change of perspective makes it trivial: imagine two monks, one going up and one coming down on the same day. They must meet. Mlodinow uses the riddle to sum up the traits of innovators: flexible thinking, patience, freedom from others' beliefs, willingness to change perspective and faith that an answer exists. Socrates compared thinking to a craft like pottery: it can be learned.
He also looks back at what the twentieth century did with the tools in this book. Understanding the chemical bond gave plastics, nylon, synthetic fertilisers and chlorinated drinking water. Biology gained DNA identification, gene splicing, optogenetics (switching neurons on and off with light in rats) and brain imaging that can, in limited cases, read simple mental states. None of it would exist without the quantum theory of Part III, which explains why atoms bond at all.
The personal note returns too. The traits that let Galileo, Darwin and Heisenberg see further are, he insists, not magic. They can be cultivated, much as his father, with no formal education beyond his teens, kept a lifelong habit of puzzling over problems. Thinking upright is a practice, not a gift.
He ends with hubris. Every age believed its picture nearly complete (the four elements, Newtonian determinism); in the 1980s Stephen Hawking predicted a theory of everything by 2000. His list of open questions: the origin of life, why sex evolved, consciousness, how water's hydrogen bonds work, how proteins fold, a Standard Model with no gravity and many fudge factors, stalled string theory, and the fact that ordinary matter is only about 5% of the universe, with the rest dark matter and dark energy (the latter inferred from the accelerating expansion discovered in 1998). He wishes he could live forever to learn the answers.
- Protein folding: AlphaFold2 predicts a protein's 3D structure from its sequence. The 2024 Nobel Prize in Chemistry went to Demis Hassabis and John Jumper for it, shared with David Baker for computational protein design.
- Dark energy: DESI DR2 (2025) reports evolving dark energy favoured over a constant at about 2.8 to 4.2σ depending on the supernova set combined with BAO+CMB; a careful combined read sits near ~3.1σ. Independent Bayesian re-analyses often land near 3.0 to 3.2σ or find no decisive preference. Strong hint under debate; short of 5σ; not a settled overthrow of ΛCDM.
- Reading minds: a 2023 study (Tang et al., Nature Neuroscience) decoded the gist of continuous language from fMRI scans of three cooperating volunteers.
- Hawking died on 14 March 2018 without his theory of everything; string theory remains experimentally unconfirmed.
- The quantum began as a reluctant fix (Planck) and became a new picture of reality (Einstein, Bohr, Heisenberg, Schrödinger).
- Progress came from anomalies taken seriously: the blackbody spectrum, the photoelectric effect, back-scattered alpha particles.
- Abandoning unobservables (Heisenberg) was as important as new mathematics.
- Quantum theory predicts probabilities; Bell tests and two recent Nobels confirm it is not a placeholder.
- Science is humbling: its open questions are as large as its answers.
Two monks and open questions
A monk walks up a mountain one day and down the next on the same path. Is there a spot he passes at the same time of day on both trips? Algebra is awkward; a change of perspective makes it trivial: imagine two monks, one ascending and one descending on the same day. They must meet. Mlodinow uses the riddle to name innovator traits: flexible thinking, patience, freedom from others' beliefs, willingness to change perspective and faith that an answer exists. Socrates compared thinking to a craft like pottery: it can be learned.
The epilogue also inventories what the twentieth century built with the book's tools. Understanding the chemical bond gave plastics, nylon, synthetic fertilisers and chlorinated water. Biology gained DNA identification, gene splicing, optogenetics and limited mind-reading via imaging. None of it works without the quantum theory of Part III, which explains why atoms bond at all.
- Protein folding: AlphaFold2 predicts structure from sequence. The 2024 Nobel Prize in Chemistry went to Demis Hassabis and John Jumper, shared with David Baker for computational protein design.
- Dark energy: DESI DR2 (2025) reports evolving dark energy favoured over a cosmological constant at about 2.8 to 4.2σ depending on which supernova set is combined with BAO+CMB; a careful combined read sits near ~3.1σ. Dataset-sensitive; short of 5σ discovery. Independent Bayesian re-analyses often land near 3.0 to 3.2σ or find no decisive preference. Treat as a strong hint under active debate, not a settled overthrow of ΛCDM.
- Reading minds: Tang et al. 2023 (Nature Neuroscience) decoded gist of continuous language from fMRI in cooperating volunteers.
- Theory of everything: Hawking died 14 March 2018 without the TOE he once hoped for by 2000; string theory remains experimentally unconfirmed.
Restate your hardest open problem so that, if the answer is yes, the meeting point is obvious. If you cannot, you may be missing a symmetry or dual view. The dark side: clever reframes that predict nothing new are rhetoric. Demand a forecast the old frame cannot make.
Score candidates (or yourself) on flexible framing, patience with anomalies, independence from fashion, willingness to re-pose and faith that tests exist. High scores without kindness still fail teams; the book assumes a community.
From bread traded for a puzzle to monks meeting on a path, the book argues that science is the most human of activities: curiosity plus culture plus measurement plus the courage to accept alien answers.
Evolving dark energy and AlphaFold are conception-and-acceptance dramas happening now. Keep the red and orange boxes as a running audit, not as spoilers of wonder.
Craft, bond and the unfinished sky
The two-monk move is a craft skill, like pottery. Socrates said thinking can be taught; Mlodinow agrees. The twentieth century's chemical bond unlocked materials and public health; biology unlocked identity and intervention; none without quantum theory. Open questions remain: proteins (now much less open thanks to AlphaFold), dark energy (DESI's evolving hint), mind-reading (early fMRI gist), and a theory of everything (still unconfirmed).
Do not assume upright habits arrive by osmosis. Run apprenticeships: anomaly triage, Wallace clocks, Halley visits, Rosetta tables, two-monk reframes. Grade on probes shipped, not on opinions held.
The book ends where it began: a species that pays bread for answers, now asking questions that outrun its senses.
Monk, traits; habits over cathedrals
Innovator habits under fog. Open questions are unfinished homework.
Name an open question and the next measurement that moves your odds by at least 10%. Schedule it.
Monk craft and open sky
The two-monk riddle is a craft drill: reframe until the answer is cheap. Socrates held that thinking can be taught; the epilogue agrees. Chemical bonding, materials and public health rode on quantum theory; biology rode on both. Open files as of October 2026: protein structure much less open (AlphaFold-class tools), dark energy with a DESI DR2 hint of evolution at roughly 2.8 to 4.2σ depending on supernova sample (still not decisive over ΛCDM in Bayesian tests), early neural decoding of gist, and a still-unconfirmed theory of everything.
Treat evolving dark energy as a live anomaly under triage, not a settled rewrite of cosmology. Chapter 10's lesson applies: instrument and model arguments first.
Two monks on one path. The meeting-point riddle yields when you change the frame: one day, two walkers. Insight is often a re-pose, not more grit on the old pose.
Restate your hardest open problem in a form that makes the answer obvious if true (the two-monk move). If you cannot, you may be missing a symmetry or a dual view.
Clever reframes can dodge evidence. A new angle that predicts nothing new is rhetoric. Demand a forecast the old frame cannot make.
Limits. Planck and 1905; prosthetic senses. Invisible. Scattering as sight; 2025 quantum-device Nobel echo. Quantum. Formalism plus diaspora plus private loss. Epilogue. Habits over cathedrals; schedule a probe.
| Move | Practice |
|---|---|
| Prosthetic sense | invent the meter |
| Scattering sight | bounce off the invisible |
| Diaspora memory | credit people |
| Open questions | next measurement moving odds ≥10% |
Part III close. When senses fail, build meters; when meters confuse, write invariants; when politics murders teachers, protect the craft and remember the names. Schedule one probe on an open question before you celebrate the last closed one.
Synthesis: Thinking Upright at Work
Mlodinow's history is a catalogue of how understanding advances: by connection, by measurement, by taking anomalies seriously and by letting go of comfortable pictures.
A useful way to read the book at work is to ask, at each episode, what made the new idea possible and what delayed it. The enablers recur: dense communities, new instruments, better notation, patrons willing to fund curiosity, and individuals stubborn enough to finish. So do the delays: reverence for authority, ideas that matched common sense too well, isolation, and the career interests of those invested in the old view. Both lists apply almost unchanged to a technology team, an architecture function or a bank's risk department.
Seven patterns from the book
- Ask why, not only how (the child with the L-shaped block). Understanding the mechanism lets you notice when it breaks.
- Connect, don't isolate (the cultural ratchet, Halley's visit). Ideas improve when reviewed; Newton's unreviewed notes stayed unfinished for 20 years.
- Measure (Oresme, Galileo, Lavoisier). Aristotle's qualitative physics lasted 2,000 years because nobody put numbers on it.
- Idealise, then add back reality (Galileo's frictionless plane). A simple model that captures the law beats a complicated one that captures nothing.
- Respect anomalies (the blackbody spectrum, alpha particles that bounced back). The small mismatch often hides the new theory.
- Abandon unobservables (Heisenberg). Build on what you can measure, not on stories about internals you cannot see.
- Expect resistance, and examine your own (Planck's principle, the two themes of Chapter 1). Incumbents, including you, defend the old picture.
AI systems
Modern AI repeats several of the book's episodes. Large models are, in Heisenberg's sense, theories of observables: they map inputs to outputs without a readable internal mechanism, which is why interpretability research exists. AlphaFold's Nobel shows a new route to discovery, prediction without full explanation, which would have troubled Einstein as much as quantum probability did.
| Book episode | AI analogue | Practice |
|---|---|---|
| Aristotle's qualitative physics | Demos judged by impressions | Build quantitative evaluation sets before shipping |
| Galileo's idealisation | Baselines and ablations | Start with a simple model; add complexity only when it beats the baseline |
| Back-scattered alpha particles | Rare failure cases | Log and study outliers; one strange output can reveal a systematic flaw |
| Heisenberg's observables | Black-box models | Judge by measured behaviour; do not trust stories about what the model "knows" |
| Uncertainty principle | Probabilistic outputs | Report confidence and calibrate it; never present a sample as a certainty |
| Newton's isolation | Unreviewed models | Peer review, red-teaming and model risk review before release |
Treat a model as a scientific hypothesis: state what it should predict, measure it against data it has not seen, and look hardest at the cases where it fails. The temptation, as with Aristotle, is to accept fluent answers because they match common sense.
Banking
Banks are measurement institutions, which makes the book's history unusually direct for them.
The institutions that thrived in the book were those that measured honestly and updated quickly. In a bank that means quantitative risk measures, independent challenge, attention to outliers and the humility to treat every model, human or machine, as provisional.
A personal checklist
- What would it predict, and how would we measure it?
- What is the simplest version that captures the core law?
- Which anomaly does it explain that the old view cannot?
- Who has reviewed it outside the team that built it?
- Which of my own beliefs am I defending because I was raised with them, like the father's roadkill deer?
- If it is right, what gap does it predict we will find next?
Worked example: a credit model through the seven patterns
- Ask why. Before accepting a lift in approval accuracy, the team asks which features drive decisions and why they should predict default. A feature that works for no causal reason is a candidate for spurious correlation.
- Connect. Credit officers, data scientists, compliance and model validation review the design together, rather than in sequence.
- Measure. Performance is measured on a held-out period that includes a downturn, by segment, with fairness metrics alongside accuracy.
- Idealise. A scorecard baseline (logistic regression) sets the bar; the complex model ships only if it beats it by a margin worth the loss of transparency.
- Respect anomalies. Monitoring flags applicants whose scores move sharply on small changes in input; each case is investigated.
- Observables. Explanations given to customers and supervisors are based on measured feature effects, not on narratives about what the model "understands."
- Resistance. The team expects pushback from officers used to judgement-based lending and from data scientists attached to their model, and agrees in advance what evidence would change the decision.
Leading change the upright way
The book's history is also a guide for anyone trying to introduce a new idea into an organisation. Galileo won the argument and lost the battle because he humiliated his patron; Darwin waited 20 years but arrived with overwhelming evidence; Planck converted because he was open to his opponents' methods; Mendeleev won because his table predicted gallium. The pattern for a new platform, model or operating model is similar: make a falsifiable prediction, gather evidence in public, give opponents a respectable path to agreement, and do not mistake being right for being persuasive.
| Idea | Proposed | Broadly accepted | Lag |
|---|---|---|---|
| Heliocentrism | 1543 (Copernicus) | Late 1600s | About 150 years |
| Oxygen chemistry | 1777 to 1789 | About 1800 | About 15 years |
| Natural selection | 1858 to 1859 | About 1870 | About 10 years |
| Light quanta | 1905 | About 1923 (Compton, Millikan) | About 18 years |
| Matrix and wave mechanics | 1925 to 1926 | About 1928 | 2 to 3 years |
The lags shrink, as the book's opening chapter predicts, because communities grew larger and better connected. The lag inside a single organisation can still be long; the cure is the same: connection, measurement and evidence.
Where to go next
- For the history of physics: Mlodinow's Euclid's Window (geometry from Euclid to strings) and Abraham Pais's Subtle Is the Lord (Einstein) extend Parts II and III.
- For the sociology of science: Planck's principle and the book's many resisted ideas are the seed of Thomas Kuhn's The Structure of Scientific Revolutions.
- For practice: apply the seven patterns to one live project this quarter, and record which anomaly you chose to investigate and what it taught you.
The book's final message is optimistic but not complacent. Each generation inherits better tools and a larger, better-connected community, so the pace of understanding keeps rising. Each also inherits the temptation to believe the picture is nearly finished. Hawking's theory of everything by 2000, Jolly's "just about complete" physics and Aristotle's nineteen centuries are warnings from the same family. The upright thinker keeps both in view: confidence that the world is comprehensible, and doubt about any particular comprehension of it.
Depth checklist for practitioners
Use this after any chapter. It compresses the upright habits into weekly practice.
- Price the question: what is the cheapest observation that would settle it?
- Name the clock: biological curiosity, cultural ratchet, civilizational tool or instrumented theory?
- Conception or acceptance: do you lack a category, or does the category threaten someone?
- Freeze interrupt: feared cause, alternative, discriminating check, then act.
- Ratchet metric: what improvement reduced toil this month, and who owns the next?
- Tablet test: who, count, date, threshold for every roadmap claim that spends money.
- Purpose versus law: rewrite teleology as a mechanism before you approve a design.
- Merton then ramp: write the mean-speed theorem, then instrument it.
- Halley visit: outsider, one question, written answer in a week.
- Gaps as predictions: two empty cells in your system table, each with a fill test.
- Wallace clock: private ideas older than twelve months must abstract, partner or die.
- Anomaly triage: instrument, operator or model before you rewrite the story.
- Observables cut: delete internal states that never meet a metric, or add the metric.
- Rosetta table: map rival languages until seven of ten concepts align.
- Two-monk reframe: re-pose until a yes would be obvious; demand a new forecast.
Science is curiosity plus culture plus measurement plus the courage to accept answers that do not look like common sense; work is the same craft under weaker incentives.
Monday curiosity invoice; Tuesday ratchet teach; Wednesday write the decision; Thursday name the invariant; Friday name the missing instrument. Shadow: praise without budget; tools without teaching; talk without record; slogans without invariants; dashboards without sensors.
Synthesis close. Upright at work is a weekly loop plus a quarterly acceptance-lag review. Tools without teaching, praise without budget and dashboards without sensors are the modern freeze. DESI hints and Göbekli debates are only vivid instances of the same habit: measure next, slogan later.
Glossary
Alpha particle. A helium nucleus (two protons, two neutrons) emitted by some radioactive atoms; Rutherford's probe of the atom.
Atomic number. The number of protons in an atom's nucleus; it, not atomic weight, orders the periodic table.
Atomic weight. The relative mass of an atom, first estimated by Dalton with hydrogen set to 1.
Bell test. An experiment comparing correlations of entangled particles with the limits allowed by local hidden-variable theories.
Blackbody radiation. Light emitted by an ideal absorber because of its temperature; its spectrum led Planck to the quantum.
Brownian motion. The random jiggling of small particles in a fluid, explained by Einstein as molecular bombardment.
Calculus. The mathematics of rates of change and accumulation, developed by Newton and Leibniz.
Conservation of mass. In a chemical reaction, total mass is unchanged (Lavoisier).
Controlled experiment. A test in which only one factor differs between groups, as in Redi's jars.
Cultural ratchet. The cumulative building of each generation on the last one's tools and knowledge.
Dark energy. The unknown cause of the universe's accelerating expansion, discovered in 1998.
Inertia. The tendency of a body to keep moving uniformly unless a force acts (Galileo, Newton's first law).
Matrix mechanics. Heisenberg, Born and Jordan's 1925 form of quantum mechanics, built from observable quantities.
Merton rule. A uniformly accelerating body covers the same distance as one moving at its average speed.
Natural selection. Darwin and Wallace's mechanism: heritable variations that aid survival and reproduction spread.
Neolithic revolution. The transition from foraging to settled farming, beginning about 12,000 years ago.
Phlogiston. A supposed substance released in burning, replaced by Lavoisier's oxidation theory.
Photoelectric effect. The emission of electrons from metal struck by light, explained by light quanta.
Photon. A quantum of light, with energy proportional to its frequency.
Quantum. The smallest discrete amount of energy exchanged, E = h × frequency.
Spontaneous generation. The disproven belief that living things arise from non-living matter.
Stationary state. In Bohr's model, an allowed electron orbit in which the atom does not radiate.
Teleology. Explanation by purpose or goal, central to Aristotle's physics.
Uncertainty principle. Position and momentum cannot both be known exactly; a limit set by nature (Heisenberg, 1927).
Wave mechanics. Schrödinger's 1926 form of quantum mechanics, based on the wave function.
About this book
The Upright Thinkers: The Human Journey from Living in Trees to Understanding the Cosmos by Leonard Mlodinow, 2015. This is a chapter-by-chapter study of the book, with summaries, maps, tables and diagrams; short quotations keep their page references. It is shared free for personal study, is not for sale, and is not endorsed by the author or publisher. Please read and buy the original.
Insight Editions · rajeshrmahapatra.com/insight-editions