Science / Physics Lesson 04

Special relativity, in one image

Moving clocks really do tick slower, and the reason fits inside a single right triangle. Here is the triangle, the experiment that forced it, the evidence that it is real, and what it does not say.

About 23 minutes
7 parts
Free lesson
Part 1 of 7
~3 min
Part 1 of 7

The most famous failed experiment

In 1887 two scientists got a wrong answer. It turned out to be one of the most useful wrong answers in science.

In the spring and summer of 1887, in the basement of a dormitory in Cleveland, Ohio, two American scientists carried out the most careful experiment of their careers. Albert Michelson and Edward Morley wanted to measure how fast the Earth moves through the aether, an invisible substance that nineteenth century physicists believed filled space and carried light, the way air carries sound. The Earth orbits the Sun at about 30 kilometres a second. If light travelled through a still aether, they reasoned, then light sent along the Earth's direction of motion should take a very slightly different time than light sent across it, like two swimmers in a river, one going upstream and back, the other across and back.

The apparatus shows how seriously they took it. Light was bounced back and forth between mirrors until its path was 11 metres long, and the whole instrument was built on a sandstone block about a foot thick and five feet square. The block floated in a circular trough of mercury, so that one gentle push would let it turn slowly through every direction while an observer looked through a telescope. The stone building and the heavy block cut out vibration. In experiments of this kind, fringes in the eyepiece could vanish because of passing horse traffic and distant thunderstorms.

The effect they were hunting was tiny, because the Earth's orbital speed is only about one hundredth of one percent of the speed of light. They calculated that it should shift the pattern by about 0.4 of a fringe, and they estimated that they could see a shift as small as 0.01. In August 1887 Michelson wrote to Lord Rayleigh that the result was "decidedly negative". The expected shift was 0.40 of a fringe, and the largest he saw was 0.02, and the average much less than 0.01, and not in the right place.

So here is the puzzle. They had found no aether wind at all. Light took the same time in every direction, at every season, however the apparatus was turned. That was the wrong answer, in the sense that nobody had a picture of light that predicted it. Wrong answers are valuable in science because they tell you that the picture in your head is broken, not just the experiment.

SHIFT IN THE LIGHT PATTERN, MEASURED IN FRINGES Expected shiftIF AETHER WIND EXISTED0.40 Smallest they could seeTHEIR OWN ESTIMATEabout 0.01 Largest shift seenNOT IN THE RIGHT PLACE0.02
From Michelson's letter of August 1887: the expected shift was 0.40 of a fringe, the largest he saw was 0.02, and the average was much less than 0.01. The experiment was sensitive enough to see an effect about forty times smaller than the one it expected.

Michelson and Morley themselves moved on. After this "failed" experiment they stopped their aether measurements and used their interferometer technique to establish the wavelength of light as a standard of length. The result stayed behind on its own. It was repeated with steadily increasing sensitivity for more than a century, and no aether wind has ever turned up.

Eighteen years later, a 26 year old named Albert Einstein was working as an assistant examiner, third class, at the Swiss patent office in Bern. He had failed to find a teaching job after university. In 1905 he published four papers, including one on special relativity. Years later he wrote that if the Michelson Morley experiment had not brought physics into serious embarrassment, no one would have regarded the theory of relativity as even a halfway redemption.

Now be careful, because the neat story is too neat. Historians disagree about how much the 1887 experiment guided Einstein. In later years he claimed it played no role in his thinking, and historians have concluded that he was at least aware of it as early as 1899. The experiment on its own also said little about whether light's speed is the same for every observer, because the source and the observer were both moving together. What it did do was make a constant speed of light easier to accept once it was proposed.

In the next six parts we will follow Einstein's two rules, build a clock out of light, find the size of the effect inside a single right triangle, look at the evidence that it is real, ask who is really moving, and then see what the same ideas say about mass and energy.

A wrong answer tells you that the picture was wrong, not that the world is.

Key idea

In 1887 the best experiment of its day found no sign of Earth moving through the aether. The result had no place in the old picture of light, and relativity grew in that gap, though historians still debate how directly it guided Einstein.

Part 2 of 7

Two rules that cannot both be true

Einstein looked at two sound rules of physics that seemed to contradict each other. He kept both.

Start with something ordinary. In the popular book Einstein wrote in 1916, he used a railway train. Suppose a train travels at 100 kilometres an hour, and a passenger walks forward along the carriage. To the passenger, their speed is whatever they walk at. To someone on the embankment, the speeds add. For a ball thrown forward at 50 kilometres an hour on that train, the person on the ground sees 100 plus 50, or 150 kilometres an hour. These are my invented numbers, but the rule is the old one, and in daily life it works perfectly.

Now state the first of Einstein's two rules. It is called the principle of relativity: the laws of nature run their course in the same way on the embankment and on a train moving smoothly along it. There is no experiment you can do inside a smoothly moving carriage that tells you it is moving. Einstein points out that this had been tested in mechanics to a standard "little short of wonderful", and he expected it to hold for all of physics.

The second rule is about light. In empty space, light travels at one fixed speed, 299,792,458 metres per second. This is the speed of light, written c, and it was known from the theory of electromagnetism developed by James Clerk Maxwell. Here comes the collision. Apply the old addition rule to a beam of light shone along the train. For the ground, the beam travels at c. For the train, the old rule would say it travels a bit slower than c. But the principle of relativity says the law of light must be the same for the carriage as for the ground. As Einstein puts it, there appears to be nothing for it but to abandon either the principle of relativity or the simple law of the propagation of light.

Most physicists of the time were more inclined to give up the principle of relativity. Einstein did something different. He asked which hidden assumption was causing the clash. Perhaps neither rule was wrong. Perhaps the old rule for adding speeds, which seemed so obviously true, depends on an assumption about time that nobody had examined.

1. A BALL THROWN FORWARD ON A MOVING TRAINtrain 100 km/hball 50ground sees 150 km/hSpeeds add.2. A BEAM OF LIGHT ON THE SAME TRAINlight: the speed c, for the trainlight: the same c, for the groundSpeeds do not add.
The bars show the idea, not a scale. If light followed the ball's rule, the ground would measure a speed a little greater than c. It never does.

Here is the assumption. Einstein says that before relativity, physics had tacitly assumed that a statement of time had an absolute meaning, independent of how the body of reference was moving. In plain words: everyone agrees what "now" means, and everyone agrees how long a minute lasts. The addition of speeds depends on this. If you drop it, the clash disappears, but the price is that time, which feels like the most solid thing there is, becomes something that depends on the observer.

Why did Einstein trust the light rule enough to give up absolute time? Partly because Maxwell's theory of electromagnetism pointed to a fixed speed for light, and partly because no evidence for an aether had turned up. There is conflicting evidence about how much the 1887 experiment influenced him, as we saw. But the rule has since been tested again and again. Experiments of the Michelson Morley type have been repeated with steadily increasing sensitivity, and the speed of light has come out the same.

It helps to say precisely what the second rule claims. It does not say light is fast. It says that if you measure the speed of a beam in empty space, you get the same number, 299,792,458 metres per second, whether you are standing still, chasing the beam or running away from it, and whether the lamp that made it is moving. Chase a light beam as fast as you like, and it still recedes from you at c.

One more piece of honesty. This lesson is about special relativity, which deals with steady motion and leaves gravity aside. A more general theory that handles gravity came a decade later, in 1915. We will meet it briefly when we get to GPS, which needs both.

Keep both rules, and something else has to give. What gives is time.

Key idea

Special relativity rests on two rules: the laws of nature are the same for every steadily moving observer, and light in empty space has the same speed for everyone. They clash only if you assume that time is the same for all, so Einstein gave up that assumption.

Part 3 of 7

A clock made of light

To see what gives, build the simplest clock imaginable and watch it from two places.

Here is a clock that no one has built, but that physicists use in their heads. Take two mirrors facing each other, a distance L apart. Fire a pulse of light from the bottom mirror. It travels up, bounces off the top mirror, and returns. Each time it comes back to the bottom mirror, the clock ticks once. For a clock a metre tall, one tick would last about 6.7 billionths of a second, which is the straight up and down path of 2 metres divided by the speed of light.

Now put this clock on a spaceship that glides steadily sideways, and watch it go by. For the crew inside, nothing looks different. The pulse goes straight up and straight down. But you are watching from outside. While the pulse is in flight, the mirrors are sliding sideways. The pulse leaves the bottom mirror, and by the time it reaches the top mirror, that mirror has moved on. So from your point of view the pulse travels along a slanted line to reach the top, and another slanted line to come back down.

A slanted path is longer than a straight one. That is geometry, nothing more. And now the second of Einstein's rules does its work. Light has the same speed c for you as for the crew. So if the pulse has farther to travel and its speed cannot change, the trip must take longer. The tick of the moving clock, as you measure it from outside, lasts longer than the tick of an identical clock at rest next to you.

AT RESTMOVING SIDEWAYSLpath is 2Llonger path, same speedD
A thought experiment, not a device anyone built at this scale. For a one metre clock at rest, one tick lasts about 6.7 billionths of a second. The grey ghosts show where the bottom mirror is at the start and the end of the tick.

Look at what has and has not happened. Nothing has broken. Light has not slowed down, because both of you agree on c. What you disagree about is how long the tick takes, because you disagree about how far the pulse went. The crew sees a short straight path, and you see a long slanted one. The speed is the same, so the times differ.

You might object that this is just a trick of one odd clock. A light clock is special, you could say, and a pendulum or a wristwatch would behave differently. But that cannot be, and the reason is the first rule. If the light clock ran slow and a pendulum clock did not, then the crew could compare them and discover that they were moving, which the principle of relativity forbids. So every kind of clock, mechanical, electronic or biological, has to slow by exactly the same factor. The light clock only reveals what time itself is doing.

With the 3, 4, 5 triangle in the next part, we will see that at 0.8 of the speed of light the slanted path is 5 thirds as long as the straight one. So the moving tick takes about 1.67 times as long. At ordinary speeds the effect is almost too small to imagine. By this arithmetic, an airliner moving at about 250 metres a second changes the rate of a clock by only about three parts in ten trillion.

One more thing to hold on to. The word moving means moving relative to the observer. For the crew, their own clock is at rest and ticks normally. There is no such thing, in this theory, as a clock that is moving all by itself. This will matter in Part 6.

A pulse of light on a slanted path has farther to go. If its speed is fixed, the tick must take longer.

Key idea

Because light has the same speed for every observer, a light clock that moves past you takes a longer slanted path, so its tick lasts longer. Since every clock must agree with it, it is time itself that runs differently.

Part 4 of 7

One right triangle

How much slower? The answer comes from a shape you met at school.

Now we measure the effect. Take half of one tick of the moving clock, from the moment the pulse leaves the bottom mirror to the moment it reaches the top. Call the length of the whole tick, as you measure it from outside, t prime. Half of it, t prime over 2, is the time the pulse is in flight upward. During that time three things happen, and they form a triangle. The clock slides sideways by its speed v times t prime over 2. That is the base. The mirrors are a fixed distance L apart. That is the height. And the pulse travels along the slanted path at speed c for a time of t prime over 2, so its length is c times t prime over 2. That is the long side.

We need one more fact, and it comes from the crew. For them the pulse goes straight up and down, so one of their ticks, which we will call t, is 2L divided by c. That means L equals c times t over 2. Now we have all three sides of a right triangle in terms of the two ticks. The base is v times t prime over 2, the height is c times t over 2, and the long side is c times t prime over 2.

Here is where Pythagoras comes in. In any right triangle, the square on the long side equals the sum of the squares on the other two. So c times t prime over 2, squared, equals v times t prime over 2, squared, plus c times t over 2, squared. Multiply it out and rearrange, and the tick of the moving clock comes out as the tick of the resting clock divided by the square root of one minus v squared over c squared. Try it with pencil and paper. It takes about four lines.

That factor has a name, the Lorentz factor, written with the Greek letter gamma. It is 1 when v is zero and grows as v approaches c. At 0.6 of the speed of light, gamma is 1.25. At 0.8 it is 1.67. At 0.9 it is 2.29. At 0.99 it is about 7.09, so that a tick that lasts one second on the ship lasts about seven seconds as measured from the ground.

vt′ / 2SIDEWAYS DISTANCELMIRROR GAP= ct / 2ct′ / 2SLANTED PATHAt 0.8 of light speed the sides are 3, 4 and 5.
Half of one tick of the moving clock. At 0.8 of the speed of light the three sides are in the ratio 3 to 4 to 5. The slanted path is 5 parts, the mirror gap is 3, so the moving tick lasts 5 over 3 of the resting tick.

Look at the picture and notice the 3, 4, 5 triangle at 0.8 of the speed of light. The mirror gap is 3 parts, the sideways distance is 4 parts, and the slanted path is 5 parts. The moving tick is therefore 5 over 3 of the resting tick, which is 1.67. That is the whole proof, in one image: a right triangle, and the fact that light has the same speed for both observers.

It is worth being clear about what the argument used. It used Pythagoras, which assumes ordinary flat geometry. It used the constant speed of light. And it used the principle that the crew sees the pulse go straight up. It did not use anything about mass, energy or force. That matters later, because this triangle gives us time dilation only. It does not by itself give us the most famous equation of all.

The factor is named after the Dutch physicist Hendrik Lorentz, who had worked out related equations before Einstein, in the framework of the aether. Einstein's contribution was to derive them from two simple rules and to say what they mean: it is time and space themselves that change, and not just the behaviour of some substance.

We should be careful about the extremes. As v gets close to c the factor grows without limit, which is part of the reason nothing is observed travelling faster than light. But what the factor does for ordinary speeds is its own kind of surprise. It is a result you can check yourself on a sheet of paper, and then check against the real world.

Pythagoras plus a fixed speed of light gives you time dilation.

Key idea

With a single right triangle and the fact that light has one speed for everyone, a moving clock's tick is longer by one over the square root of one minus v squared over c squared. At 0.8 of the speed of light, the triangle is 3, 4, 5 and the tick is 5 over 3 as long.

Part 5 of 7

Is it real?

A clever argument is not evidence. What have we actually measured?

It is a surprising piece of history that special relativity was not tested for time dilation at once. The Nobel committee, which gave Einstein the 1921 prize for his work on the photoelectric effect and not for relativity, is said to have waited patiently for experimental confirmation of special relativity, and none came until the time dilation experiments of Ives and Stilwell in 1938 and 1941, and of Rossi and Hall in 1941. The theory had been argued about for over thirty years before it was directly checked.

The Rossi and Hall experiment used particles called muons, which are made when cosmic rays strike the upper atmosphere. A muon at rest lives only about 2.2 millionths of a second before it decays. At nearly the speed of light that is only about 660 metres of travel, far short of the many kilometres from the upper atmosphere to the ground, unless its clock runs slow. A muon moving at 98 percent of the speed of light has a lifetime about five times longer, in agreement with what is observed. Rossi and Hall compared the number of muons at the top of a mountain with the number at sea level.

The effect has been measured far more precisely since. In a muon storage ring at CERN, muons circulating with a Lorentz factor of 29.327 were found to live 64.378 millionths of a second, which agrees with the prediction to within about nine parts in ten thousand, give or take four. The decay of a muon is a clock, and it ticks slower when the muon moves fast.

Then there is the airliner experiment. In 1971 the physicist Joseph Hafele and the astronomer Richard Keating took four caesium atomic clocks on commercial flights twice around the world, first eastward and then westward. They compared the clocks with ones kept at the United States Naval Observatory. When the clocks were reunited, the three sets disagreed with one another, in amounts that matched the predictions of special and general relativity.

CLOCK DIFFERENCE AFTER THE TRIP, IN BILLIONTHS OF A SECONDEastwardPREDICTED−40EastwardMEASURED−59WestwardPREDICTED+275WestwardMEASURED+273GAINED (RIGHT) OR LOST (LEFT) COMPARED WITH CLOCKS ON THE GROUND
Hafele and Keating, 1971, published in 1972. Each result carries an error: eastward measured 59 lost, give or take 10, against 40 predicted, give or take 23. Westward measured 273 gained, give or take 7, against 275 predicted, give or take 21. The effect of speed and the effect of altitude are both included.

Look at the numbers in the chart. The eastward flight lost about 59 billionths of a second against a prediction of 40, and the westward flight gained about 273 billionths against a prediction of 275. Notice the error bars in the caption. This was a small experiment, with only four clocks, and its precision was about 10 percent, so it is better read as a confirmation of the sign and size of the effect than as a precise test. Two effects are mixed in it: the slowing from speed, which is what we derived, and a speeding up from altitude, which comes from general relativity.

The effect is in your pocket. The GPS satellites carry atomic clocks and orbit at high speed and altitude. The system builds in two relativistic corrections, one for the relative speed of satellite and receiver, and one for the difference in gravity. By standard estimates the satellite clocks, left alone, would run fast by about 38 millionths of a second a day, which is about 45 gained from weaker gravity and about 7 lost from speed. Turned into distance, that is an error of the order of ten kilometres a day.

Closer to home, an astronaut who spends six months on the International Space Station, moving at about 7,700 metres a second, ages about 5 milliseconds less than a person on the ground. The cosmonaut Sergei Krikalev, after a long career in orbit, experienced about 20 milliseconds of difference. These numbers are tiny, which is why you never notice them, and measurable, which is why we know they are real.

It is fair to say what this does and does not show. It shows that a formula we derived from a triangle matches clocks and particles over a huge range of speeds, and that it has passed tests of steadily increasing sensitivity. No experiment can prove a theory forever. What it can do is fail to break it, again and again, and this one has not broken yet.

The theory waited decades for its first direct test, and has passed every test since.

Key idea

Muons, airliner clocks and satellite clocks all show moving clocks running slow by the amount the triangle predicts. The first direct tests came in 1938 and 1941, more than thirty years after the theory, and the effect is now built into GPS.

Part 6 of 7

Who is moving? Both of them

If a moving clock runs slow, who is the one moving? The answer is stranger than you would expect.

Here is a puzzle that has bothered students for a century. A spaceship passes you. You measure its clock, and by the rule we derived, it runs slow. But the crew look out of the window and see you flying past them. By the same rule, they measure your clock as the slow one. Each of you says the other's clock runs slow. Surely they cannot both be right?

They can, and the reason is that what each of you measures is different. There is an everyday analogy. If two people stand far apart, each looks small to the other, and there is no contradiction, because we know about perspective. Time dilation between steadily moving observers is reciprocal in the same way. The surprise is that nothing in physics picks one of you as the one who is really moving. That is exactly what the principle of relativity said, and it is what the Michelson Morley result looked like from the inside.

The deeper reason is something Einstein explains with lightning. Imagine two bolts of lightning that strike the railway embankment at two far apart places, A and B. An observer on the embankment standing exactly halfway between them sees the two flashes at the same moment, and says they were simultaneous. Now consider a passenger on a fast train who is exactly halfway between A and B at the instant of the flashes. The train carries her toward B and away from A, so the light from B reaches her first. She concludes that B struck earlier.

Einstein states the result in his book. In his words: "Events which are simultaneous with reference to the embankment are not simultaneous with respect to the train, and vice versa." And a little later: "Every reference body has its own particular time; unless we are told the reference body to which the statement of time refers, there is no meaning in a statement of the time of an event." This is the heart of it. The two observers disagree about which events happen at the same time, so when they compare clocks, they are comparing different pairs of events.

Clock Aon the groundClock Bin the shipA: B runs slowB: A runs slowTHE TWIN WHO TURNS AROUND IS DIFFERENTGoes outone frameTurns aroundchanges frameComes backa second frameOn reunion the traveller is younger.
While they move steadily, each observer measures the other's clock as slow, and there is no contradiction. The twin who turns around is not in the same situation as the twin who stays, so their comparison on reunion has one answer.

What about the twins? Suppose one twin stays on Earth and the other flies off at high speed, turns around and comes back. By the reciprocity rule, it sounds as though each should find the other younger, which is impossible when they stand side by side again. The way out is that the situation is not symmetric. The twin who stays at home remains in one steady frame the whole time. The traveller is in one frame on the way out and in a different frame on the way back, and the turn is where it changes. When they meet, there is one answer, and it is that the traveller is younger.

This is not only a story. The airliner clocks of 1971 were a small version of it: clocks that went round the world and came back showed real differences from clocks that stayed. And the muons in the storage ring are travellers that go round in circles, which is a constant change of direction, and they too outlive their stay at home siblings.

There is a lesson about how to think here. When the theory seems to give two contradictory answers, check what is being compared. Most apparent paradoxes in relativity come from comparing events as though everyone agreed on when they happened. Once you ask what each observer actually measures, using signals travelling at the speed of light, the contradictions go away.

Be honest about the difficulty. This is the part of relativity most people find hardest, and it is fine to need to read it twice. The best test of understanding is not memorising the words but being able to explain why the train passenger sees B first, and what that does to the word simultaneous.

There is no answer to who is really moving, and the theory works because it does not need one.

Key idea

Each steadily moving observer measures the other's clock as slow, because they disagree about which events are simultaneous. The twin who turns around changes frames, which is why that twin comes back younger.

Part 7 of 7

What E = mc² says, and what it does not

The triangle gave us time. The most famous equation in science comes from the same two rules by a different route.

The equation everyone knows is not in Einstein's relativity paper. Later in 1905 he published a short follow up with a question as its title: does the inertia of an object depend upon its energy content? In it he does not write E equals mc squared. He states a narrower result: if a body gives off an amount of energy L by emitting light, its mass decreases by L divided by c squared. It was the relation between a change in mass and a change in energy, and from it he concluded that mass and energy are two names for the same underlying conserved quantity.

Read that as a recipe for converting. The speed of light, squared, is an enormous number, 89,875,517,873,681,764 in metres squared per second squared. So a tiny amount of mass corresponds to an enormous amount of energy. By this arithmetic, one gram of matter, fully converted, would release about 90 trillion joules, which is about the energy of 21 kilotons of TNT. Almost nothing ever converts completely, but this is the exchange rate.

The Sun is the great example. In its core, four hydrogen nuclei fuse into one helium nucleus, and the helium has slightly less mass than the four hydrogen nuclei together. About 0.7 percent of the fused mass leaves as energy. Added up over the whole Sun, that is a conversion of about 4.26 billion kilograms of mass into energy every second, and that energy is sunlight. A star shines because it is losing mass.

Now, a point about honesty in teaching. The right triangle in the last four parts does not prove this. The triangle gave us time dilation from the constant speed of light and Pythagoras. The mass and energy result needs a different argument, which Einstein gave in the short 1905 paper, though it rests on the same two rules. Part of understanding relativity is knowing which result came from which argument.

FOUR HYDROGEN NUCLEI BECOME ONE HELIUM NUCLEUS99.3% stays as mass0.7% leaves as energy4.26 billionkilograms of the Sun's mass turned into energy, every second
Figures from standard accounts of the Sun. The tiny gold slice is what makes sunlight. Because the speed of light is so large, a tiny amount of mass corresponds to an enormous amount of energy.

Why do we not notice this in daily life? Einstein answers in his own book, where he writes that the energy changes we can produce in a system are not large enough to show up as a perceptible change in its mass. When you burn a log, the ash and gases weigh slightly less than the log and the oxygen, but the difference is far too small to weigh. Only in nuclear processes, where a larger fraction of the mass is converted, does the change become large enough to matter.

Let us put the whole lesson into one breath, because that is the test. In 1887, a careful experiment found no sign of motion through the aether. Einstein kept two rules, that the laws of nature are the same for every steady observer and that light has the same speed for everyone, and gave up the assumption that time is the same for all. A light clock then shows that a moving clock must tick more slowly, by a factor the Pythagorean theorem gives. Muons, airliner clocks and satellites bear that out. Each observer sees the other's clock slow, because they disagree about simultaneity, and mass and energy turn out to be two faces of one thing.

It helps to correct a common misreading. The name relativity makes people think that everything is relative, and that is not what the theory says. The speed of light is the same for everyone, and the laws of physics are the same for everyone. What changes is time and distance. Another way to put it: the theory is about what stays the same when you change your point of view.

And to be clear about the limits. Special relativity leaves gravity out. The theory of gravity that Einstein completed a decade later, general relativity, is needed alongside it for GPS and for much else. There are questions about how Einstein arrived at his ideas that historians still debate, and there are experiments still being done that test these ideas to greater and greater precision. That is how a good theory behaves: it keeps inviting you to try to break it.

A star shines because it is turning mass into light.

Key idea

Einstein’s 1905 follow up paper showed that when a body gives off energy, its mass falls by that energy divided by c squared. Mass and energy are two names for one conserved quantity. The triangle explains time. A separate argument, from the same two rules, gives mass and energy.

The lesson
What you can now explain
  • What Michelson and Morley expected, what they found, and why a wrong answer was valuable.
  • Einstein's two rules, and the hidden assumption about time that let him keep both.
  • How a clock made of light shows that a moving clock must tick slower.
  • Where the size of the effect comes from: one right triangle, the 3, 4, 5 case, and the Lorentz factor.
  • The evidence: muons, airliner clocks, satellites, and why the first direct tests took decades.
  • Why each observer sees the other's clock slow, what changes for the twin who turns around, and what the triangle does not prove.

Where these facts come from. Albert Einstein, Relativity: The Special and General Theory, written in 1916 (English translation by Robert W. Lawson, 1920), for the train and embankment examples, the clash between the two rules and the sentences on simultaneity. Michelson's letter to Lord Rayleigh, August 1887, and the account of the experiment in standard histories. The derivation of time dilation from a light clock, the muon, Ives and Stilwell, Rossi and Hall and CERN storage ring results, the Hafele and Keating results (published in Science, 1972), the GPS corrections and the figures for the International Space Station are from standard reference accounts. The solar figures are from standard accounts of the Sun. Quotations have hyphens and dashes removed.

Where the evidence is uncertain. Historians disagree about how much the 1887 experiment influenced Einstein, who later said it played no role, though he was aware of it by about 1899. The Hafele and Keating experiment had a precision of about 10 percent. The GPS figures (about 38 millionths of a second a day, about ten kilometres of error) come from secondary summaries of the standard accounts. The one gram figure is straight arithmetic from c squared.

This lesson explains a scientific theory. It is not a substitute for a physics course, and the full mathematics is more involved than the light clock suggests.

Continue reading

New lessons,
every week.

One lesson a week. Free while we build.

Free during early access · No spam