The Story of Time

We have built instruments that can measure a second to a precision that would not drift across the age of the universe. We still cannot tell you what a second is.

Observation

There is a moment in the eleventh book of Augustine's Confessions, written around the year four hundred in a North African port city, where the bishop stops his argument and admits something that does not fit the man who wrote the rest of the book. He has just spent ten books explaining, with a confidence that still unsettles the reader sixteen hundred years later, his sins and his conversion, the nature of memory, the soul's ascent toward God. Then he gets to time and the machinery stops. If no one asks me, I know. If I try to explain, I do not.

Fifteen hundred years before Augustine sat down to write that confession, another voice had already spoken about time, and it had not admitted confusion at all. On a battlefield in northern India, a warrior named Arjuna had lost his nerve at the prospect of fighting his own kinsmen, and his charioteer, who turned out to be the god Krishna, had revealed his cosmic form to settle the question. What Arjuna saw was overwhelming — a vision that contained all worlds and all ages — and in the middle of it Krishna told him plainly who he really was. I am Time, he said. The destroyer of worlds. I have come to gather all of you in. These warriors you do not want to kill are already dead. Your refusal to fight will not save them. Nothing will.

These two voices are the essay's whole difficulty held in a single frame. On one side, a bishop in North Africa confessing that time is the thing he cannot explain even though he knows exactly what it is. On the other, a god on a battlefield declaring that time is precisely what he is, and that the killing has already happened in the longer view of things. The declaration and the confusion sit side by side, and between them lies every human attempt since to understand what we are held inside. We are going to walk through some of those attempts now — through clocks and equations and the most precise measurements humans have ever built — and at the end of the walk we are going to find that Augustine's confession and Krishna's declaration have not been resolved by any of it. They have only been dignified.

The admission Augustine makes is disarming precisely because Augustine is not a man prone to admitting confusion. And what he is confessing is something every reader of the sentence immediately recognises as true. You know what time is. You know it when a week of grief stretches to the length of a month, and a year of routine collapses into nothing in memory. You know it when you wake from a dream and cannot place how long you have been away. You know it when you are suddenly older than your parents were when you first thought of them as old. You know it the way you know hunger or hope — not as something you have analysed but as something you have lived inside. And then someone asks you to explain it, and the thing you knew a moment ago has evaporated.

This confusion is very old. It has not been resolved. It has only been measured more precisely.

Exploration

The long count

Long before anyone tried to define time, people tried to count it. The oldest surviving instrument for tracking its passage is an Egyptian alabaster water vessel, excavated at the Temple of Amen-Re at Karnak in 1904, dating from the reign of Amenhotep the Third more than three thousand years ago. Before that, shadow clocks. Before them, notched bones from the Palaeolithic — tally marks for lunar cycles carved by people who had no written language but already knew that the moon came back. Every human culture, without exception, has invented some device for counting the days. The Aleut calendar named March when straps are eaten because by then the winter food was gone. The Wathaurung messengers in Australia tracked a fourteen-day journey by painting mud stripes on their arms and erasing one each dawn. The Chinese astronomer Su Sung built a water-driven astronomical clock in 1088 with mechanical manikins that struck gongs to mark the hours. The instruments vary. The compulsion does not. It is one of the most reliable things about being human — that we cannot stop trying to mark the passing of something we cannot name.

The instruments got steadily better. Verge escapements in thirteenth-century European monasteries, accurate to perhaps fifteen minutes a day. Christiaan Huygens' pendulum clock in 1656, accurate to fifteen seconds. John Harrison's marine chronometer, which in 1761 crossed the Atlantic to Jamaica and lost five seconds in eighty-one days, and with it solved the longitude problem that had drowned a generation of British sailors. Quartz oscillators in 1927. The first cesium atomic clock at the National Physical Laboratory in Teddington in 1955, whose inventor Louis Essen is said to have told his director they were there to witness the death of the astronomical second and the birth of atomic time. He was right. In 1967 the definition of the second itself was handed over to the atom: one second became the duration of exactly 9,192,631,770 oscillations of radiation emitted by a cesium-133 atom moving between two specific energy states. The sun and the planets, which had told humans what an hour was for every moment of recorded history, were quietly retired from the job. The most recent strontium optical lattice clocks at JILA in Colorado have a systematic uncertainty so small that if one had been running since the Big Bang, it would not have gained or lost a second by now.

We have built an instrument that would have kept perfect time across the entire age of the universe. And we still cannot tell you what it is keeping time of.

Part of the trouble is that we do not experience time with a clean organ the way we experience light with an eye or sound with an ear. The best current neuroscience locates our sense of duration somewhere in the parietal cortex, in a region called the intraparietal sulcus, where the same neural substrate handles time, space, and number together — a clue that for the primate brain these three categories are entangled from the start. Our circadian rhythms keep us roughly coupled to the planet's rotation, but the coupling is sloppy. An hour as felt by the human brain runs, on average, slightly long. When we sit quietly and try to tell how long a minute has been, we overshoot. When we are bored we overshoot more. When we are in love or in danger we undershoot. The organ that is supposed to keep us honest about time is also the organ that lies most about it, and the lies are not small. Augustine did not have a neuroscience lab, but he had noticed. A creature whose brain has no clean circuit for the thing it is trying to describe will, of course, know the answer until someone asks for it.

Newton's clock and Einstein's rebuke

For most of the Western intellectual tradition, the confusion Augustine described was understood as a limitation of human experience, not a limitation of the subject. Behind the muddled inward sense of duration, it was assumed, there lay a clean and orderly thing — time itself — ticking onward without reference to anything human. Isaac Newton gave this assumption its sharpest form in 1687, in the opening pages of the Principia, where he declared that absolute time flowed equably without relation to anything external. The sentence is doing a great deal of work. Newton is saying that time is a real, independent feature of the universe, a stage on which events happen, a parameter that would tick even if nothing existed to be ticked at. His great rival Gottfried Leibniz disagreed, arguing in a famous correspondence with the English theologian Samuel Clarke that time was nothing more than the order in which events succeed one another — a relational concept, not a substance — but Leibniz's view was the minority report for the next two centuries. Newton's clock won because Newton's clock worked. His equations predicted the motion of the moon and the return of comets with an accuracy nothing else came close to. A universe in which time is an absolute background turned out to be a universe that was much easier to calculate.

Then, in the summer of 1905, a twenty-six-year-old patent clerk in Bern published a paper that took the Newtonian clock apart. Albert Einstein's special theory of relativity began with two postulates — that the laws of physics are the same for all observers moving at constant velocity, and that the speed of light is the same for all of them — and ended somewhere nobody had expected. If both postulates are true, then two observers moving relative to each other cannot agree on how long a given event takes. A clock that moves past you at high speed ticks more slowly than a clock at rest beside you. Not because of mechanical interference, but because the thing the clocks are measuring has itself come apart. There is no single universal duration for any event. There is only the duration as measured by a particular clock on a particular worldline. A decade later, in his general theory of relativity, Einstein showed that gravity does the same thing. A clock in a strong gravitational field ticks more slowly than a clock higher up. Not slightly, not metaphorically, but literally and measurably. The Newtonian stage on which events happen turned out not to exist. There was only the four-dimensional fabric of spacetime, and every clock's reading depended on its path through it.

The easy thing to say about Einstein's theory is that it is strange. The harder and more important thing to say is that it is correct, and that we know it is correct because we have measured it with increasing precision for more than sixty years. In 1959, at the Jefferson Physical Laboratory at Harvard, Robert Pound and Glen Rebka set up an experiment in a twenty-two-metre tower and showed that gamma rays climbing out of Earth's gravitational well lost energy in exactly the amount general relativity predicted — the first laboratory confirmation of gravitational time dilation. In 1971, Joseph Hafele and Richard Keating flew four cesium atomic clocks around the world on ordinary commercial jets, eastward and westward, and compared them with reference clocks at the US Naval Observatory. The moving clocks disagreed with the stationary ones in precisely the direction and amount Einstein's equations required. The experiment cost the Office of Naval Research roughly eight thousand dollars. It is still one of the most elegant proofs of the twentieth century.

Then the measurements got more precise, and then they got almost absurd. The global positioning system in the phone in your pocket relies on atomic clocks aboard satellites orbiting roughly twenty thousand kilometres above Earth, moving at about fourteen thousand kilometres an hour. Those clocks experience two relativistic effects at once: they run slower than ground clocks because they move faster, and they run faster than ground clocks because they sit higher in Earth's gravitational well. The net effect is that each satellite clock gains about thirty-eight microseconds a day relative to the clock on your kitchen counter. It is a tiny number. But GPS position is derived from the time difference between signals arriving from several satellites, and a thirty-eight-microsecond daily drift, uncorrected, would translate into a positioning error of about ten kilometres within twenty-four hours. Every GPS satellite has its clock frequency pre-adjusted at the factory before launch, so that when the two relativistic effects begin acting on it in orbit, it will tick at the same rate a ground clock does. This correction is not an optional refinement. It is the reason your phone's map works. Einstein's strange claim about time is engineered into the navigation system of every car and every aeroplane on the planet, quietly, all the time, without you ever noticing.

In 2010, a team at the National Institute of Standards and Technology in Boulder compared two aluminium-ion optical clocks connected by a seventy-five-metre fibre-optic cable and showed that raising one of the clocks by just thirty-three centimetres — about the height of a step — produced a measurable difference in how fast the two clocks ticked, exactly in line with general relativity. In 2022, Jun Ye's group at JILA pushed the measurement further still. Using roughly a hundred thousand ultracold strontium atoms held in an optical lattice, they detected gravitational time dilation across a sample just one millimetre tall. A millimetre. The width of a grain of rice. Across that distance, time measurably runs at different rates, and we can see it.

This is the part we know. Time is not an absolute background. It is a local quantity that depends on where you are and how fast you are moving. Every ancient intuition about a universal now, a single cosmic second ticking the same everywhere at once, is simply wrong. Not philosophically wrong. Measurably, engineered-into-your-phone wrong. Augustine could not have known any of this. And yet it does not answer his question. It only changes which part is confusing.

The disagreement

Because once the Newtonian stage has been dismantled, a much harder question opens up underneath. If time is not a fundamental background parameter, what is it? A property of the fabric of spacetime, woven inseparably with the three dimensions of space? An emergent phenomenon, the way temperature emerges from the jostling of molecules — not fundamental in itself, but statistical? An illusion produced by the peculiar way our memories are structured? Or perhaps the most fundamental thing of all, more basic even than the laws of physics?

The honest answer, in the year 2026, is that the people who know the most about this disagree with each other in ways that cannot be reconciled by gathering more data. Carlo Rovelli, one of the founders of loop quantum gravity, has spent the last decade arguing that time at the fundamental level does not exist — that what we experience as the flow of time is a statistical artefact of our blurred, macroscopic view of a timeless quantum substrate. In his formulation, the universe is not made of things but of events, relations, occurrences that happen and leave traces. There are no things, only events, he has written. The difference between past and future, for Rovelli, reduces entirely to the fact that the early universe had very low entropy, and the whole of our experience of time's passage is a consequence of thermodynamics. Julian Barbour, a British physicist who spent most of his career outside the academy, goes further still. Time, he argues, is not merely emergent but simply does not exist at all — not even as an illusion. The universe is a collection of static configurations, timeless moments arranged in what he calls Platonia, and our sense that we are moving through them is a trick played by the way memory is stitched into each moment.

Lee Smolin, who helped Rovelli found loop quantum gravity in the nineteen-eighties, thinks they are both profoundly wrong. Time, Smolin argues, is the most fundamental feature of the universe — more basic than the laws of physics themselves, which he believes actually evolve over cosmological history. Space may be an illusion, but time must be real. The physicist George Ellis, in Cape Town, holds yet another view: that the past is real, the future does not yet exist, and the present is a continually growing edge — a crystallising block universe, neither Newton's frozen eternity nor Barbour's collection of static frames. The American philosopher Tim Maudlin, approaching the question from the philosophy of physics, argues that time's passage is primitive and irreducible — that every attempt to explain it away in terms of entropy or memory simply presupposes what it is trying to explain.

These are not the opinions of cranks. They are the considered views of some of the most careful minds currently working on the question, people who understand one another's arguments in technical detail and remain in genuine disagreement after decades of conversation. Einstein himself, writing a letter of condolence to the family of his lifelong friend Michele Besso in the spring of 1955, a few weeks before his own death, remarked that for physicists the distinction between past, present and future is only a stubbornly persistent illusion. The line has been quoted many times and often misunderstood. Rovelli has pointed out that it appears in a letter of consolation, not in a physics paper. Maudlin has pointed out that nothing in relativity, properly interpreted, actually requires that reading. The man who gave us our best theory of time seems to have thought it was unreal, but even that is contested.

This is where physics leaves us. We have measured time so precisely that our instruments can detect it flowing at different rates across a millimetre. We have engineered its relativity into the satellites that know where we are. And we have no consensus at all about what the thing we have measured actually is.

Kala — the destroyer and the teacher

Long before Augustine sat down to puzzle at time in Latin, and many centuries before Einstein dismantled Newton's clock, the question we are tracing was asked in Sanskrit, and it was given an answer that modern physics has never quite matched for clarity or courage. The answer lay hidden inside the Sanskrit word for time itself.

The word is kala. And kala does not only mean time. It also means death. Not two related concepts that happen to share a root — one word carrying both meanings at once. The Puranic tradition did not need a separate vocabulary for mortality because mortality, once you looked at it honestly, was what time was at its deepest level. Everything that exists in time must end; time is therefore the real agent of every ending; and so time and death collapse into a single divine function. Kala personified is Yama, the lord of death. The Bhagavata Purana describes time as a serpent that moves silently, devouring all, including even Brahma, the creator of worlds. Shiva in his fiercest aspects carries the epithets Mahakalabhairava — the one of whom even Time is afraid — and Kalasamhara, the destroyer of Death. These are not separate deities. They are the same insight expressed in iconography. The thing that takes everything from us is not outside us. It is the current we are held inside.

The Bhagavata Purana tells a story about this, in its ninth canto, that is worth telling carefully. A king named Kakudmi takes his daughter Revati to Brahmaloka, the realm of the creator, to ask Brahma for advice about whom she should marry. They arrive while Brahma is listening to a performance by the celestial musicians, and they wait politely for the music to end. When Kakudmi finally presents his question and begins listing the young princes he has in mind for his daughter, Brahma interrupts him gently. Twenty-seven chatur-yugas have passed on Earth, he tells the king, during the time you have been standing here. The young men you are considering are gone. So are their sons and their grandsons and the names of their clans. You have been waiting for what seemed like the length of a song. More than a hundred million years of Earth time have gone by. Kakudmi returns to a world he does not recognise.

The story is usually read today as a charming prefiguration of special relativity, and there are writers who will tell you excitedly that the ancient rishis knew about time dilation. This is the wrong reading, and it shrinks both sides of the comparison. The rishis did not know about time dilation in the sense we know it now; they did not have clocks at the millimetre scale and they did not derive the Lorentz transformation. What they had was something more interesting. They had an imagination of time as a presence rather than a parameter — as something that could flow differently in different regions because time itself was a divine potency, not a uniform backdrop. Modern physics has confirmed that time can flow differently in different regions. It has not confirmed the Puranic framing of why. Perhaps it never will. The resonance between the two pictures is real, but it is a resonance, not an identity, and the essay is better for respecting the distance.

What the Puranic tradition did have, and what modern physics has not offered and probably cannot offer, was an answer to the question of how a creature should live inside time. The answer begins with the destroyer. Time is the serpent; Kala will take everything; the king who goes to the court of Brahma will come home to find his entire world gone. This is not consolation. It is the blunt first step. But then the tradition makes a second move that is easy to miss and absolutely essential. The very destructiveness of time, it insists, is what makes time the mahaguru — the great teacher. Because Kala will take everything, the moments we have acquire their entire weight from that fact. A universe in which we did not die would be a universe in which nothing mattered, because nothing would have to be chosen now rather than later. The Bhagavata's instruction to waste no moment is not productivity advice. It is a recognition that seriousness — the seriousness of commitment, of love, of work, of care — depends entirely on the awareness that the moment will not come again. Time destroys; the awareness of that destruction, faced rather than fled, teaches us how to be.

The thing that will take everything from us is also the thing that makes anything matter at all. The destroyer is the teacher. You cannot have one without the other.

What is striking is that this is not a particularly Hindu idea. It is what thoughtful observers keep arriving at when they look carefully at the same question from different cultural vocabularies. In 1927, the German philosopher Martin Heidegger published a book called Sein und Zeit — Being and Time — in which he argued that death is not primarily an event at the end of a life but an ever-present possibility that structures the whole of existence. He called this being-toward-death. Most of us, he said, spend our days fleeing this awareness, treating death as something that happens to other people, borrowing our lives from the anonymous opinions of the crowd. Authentic existence requires what he called resolute acceptance of one's own finitude — not morbid obsession but sober acknowledgment that time is limited and cannot be bypassed. And when that acknowledgment arrives, time stops being a sequence of clock units and becomes the horizon within which our possibilities show up as urgent, finite choices. Because we will die, Heidegger argued, our time is not merely passing — it is charged. That is why our choices matter at all.

The parallel with the Puranic tradition is exact enough to be unsettling. A Sanskrit text compiled over many centuries by forgotten rishis, and a twentieth-century German phenomenologist working in a Catholic philosophical tradition, arrive at the same observation from opposite ends of the world. If you try to refuse the destroyer you also refuse the teaching, and you drift through your days in what Heidegger called inauthenticity and what the Bhagavata would call pramada — the spiritual carelessness that mistakes the urgent for the important and forgets that the moment will not come again.

In our own decade, the American psychiatrist Irvin Yalom, working in the existential therapy tradition that descends from Heidegger through Karl Jaspers and Viktor Frankl, has documented the same observation clinically. Patients who consciously work with their mortality — who let the fact of their death sit in the room with them rather than push it out — typically report, over time, a clarifying of priorities, a deepening of relationships, and a greater willingness to live in alignment with their own values rather than borrowed ones. Patients who flee the awareness typically present instead with what the literature calls mortality salience defences: clinging harder to status, to group identity, to symbolic immortality, to the distractions that let them not notice. The destroyer does not go away because you refuse to look. It only loses the power to teach.

Ground

So where does this leave us? We have walked a long path. From Augustine's admission in a fifth-century port city, through thirty centuries of counting and clock-making, through the shattering of Newton's absolute stage and the experimental confirmations of Einstein's strange replacement, through the live disagreement among Rovelli and Smolin and Barbour about what it is we have actually been measuring, and finally through the older and quieter observation that the rishis and Heidegger and Yalom all arrived at from different directions. What is the honest resting place?

Begin with what we know, because there is a great deal of it, and it deserves plain acknowledgment. Time is real enough to be engineered around. It dilates with velocity and with gravity, and the dilation is not a philosophical flourish but a thirty-eight-microseconds-per-day correction on every GPS satellite orbiting above your head, without which the map on your phone would wander by ten kilometres in a day. We have measured time flowing at different rates across a single millimetre of space. We have built clocks so steady that the best of them would not have gained or lost a second across the entire age of the universe. The precision is astonishing. It is the honest fruit of several centuries of patient work, and it should not be waved away.

And yet with all that precision, we still cannot tell you what time is. Rovelli says it emerges from a timeless quantum substrate. Smolin says it is the most fundamental feature of reality. Barbour says it does not exist at all. Ellis says it is continually crystallising into being. Maudlin says its passage is primitive and irreducible. These are not fringe positions — they are the considered views of serious physicists who understand one another's arguments perfectly and who, after decades of conversation, remain in genuine disagreement. Augustine's confession in the year four hundred has not been superseded by any of this. It has only been dignified. If no one asks us, we know what time is. If someone asks us to explain, we find that the best minds of three civilisations, working across three millennia, arrive at the same admission from different angles.

But here is the thing the essay has been circling toward, and it is the part that matters. We do not have the luxury of waiting for the question to be resolved before we live our lives. Every morning we wake up inside a process we cannot define, and we must act anyway. We must raise our children and bury our parents. We must do our work and keep our promises. We must decide whether to speak or stay silent, whether to forgive or refuse, whether to stay in a job or leave it, whether to love the people in front of us or hold ourselves at a distance from them. None of these decisions can be postponed until the physicists agree on whether Rovelli or Smolin is correct. The metaphysics of time may remain unresolved for another thousand years. Our own small share of it is running out right now, as you read this sentence and as I write it.

This is the quiet place where the rishis and the phenomenologists were standing all along. They were not asking what time is. They were asking how a creature should live inside it, given that it will take everything. Their answer does not depend on whether time is fundamental or emergent. It is addressed to the practical fact of being a mortal thing that must choose. Face the destroyer, they said, and let it teach you. Recognise that every moment acquires its whole weight from the fact that Kala will take it. Waste nothing. Do not confuse the urgent with the important. Do not borrow your life from the opinions of people who have already forgotten they are mortal. Act, within the dharma you have been given, without attachment to the particular outcome, because the outcome is not finally yours to command. You are not outside Kala. You are a small thing held inside it. And precisely for that reason, every act you perform carries a seriousness that would vanish if you were eternal.

Remember the voice that opened this essay. On a battlefield in northern India, at the moment when a warrior asked to see the true nature of what he was held inside, the god at the centre of the chariot revealed himself as Time — the destroyer of worlds, the gatherer-in of all things — and told Arjuna that the killing had already happened in the longer view of things. The revelation did not resolve Arjuna's doubt. It recontextualised it. The warriors on both sides of the field were already dead, Krishna said; Arjuna's task was not to decide the outcome but to recognise his place inside a process larger than his will and to act, in the role that was his to perform, with whatever clarity and courage he could find. The same voice that declared itself the destroyer told Arjuna, a few chapters earlier, exactly how such a creature should act inside the current that will take him. The instruction is older than physics and it has not been improved upon. The action is required anyway.

That is the honest end of this essay. We do not know what time is. We may never know. The best science we have ever built can measure its local behaviour to astonishing precision and remains silent about its deeper nature. The best philosophy we have ever written can name the confusion beautifully and cannot dissolve it. The oldest traditions we have inherited do not answer the question of what time is either, and they knew they were not answering it. What they answer is the only question we actually have to answer in order to live — which is how to be a creature held inside a current we did not choose, moving toward an ending we cannot postpone, in a universe whose deepest nature may be permanently hidden from us. Their answer is not a solution. It is a posture. Recognise the destroyer. Let it teach. Do your work. Love the people in front of you. Waste no moment. And when the current takes you, as it will, be the kind of person who met it honestly rather than the kind who looked away.

That is all any of us can do. It turns out to be enough.