The Evidence and the Everyday
Two phases of thought experiments. Now the fair question: is any of this true, or is it a beautiful story physicists tell each other? This phase is the answer, and the answer is that relativity is among the most tested, most confirmed ideas in all of science. We'll start with the proof you carry in your pocket, then look up at bent starlight and a wobbling planet, and finish with the day humanity heard two black holes collide.
A promise about honesty: science is built on what survives testing, so we'll be clear about what's rock-solid and what's still being refined. The headline is that the core of relativity has passed every test thrown at it for over a century.
GPS: relativity, every second, in your hand
Your phone finds its location by listening to a fleet of satellites, each carrying an atomic clock and broadcasting the time. Your phone compares the times from several satellites and works out how far each signal traveled — and from those distances, where you are. The whole system is a race between light signals, so it lives or dies on the clocks agreeing. And here's the thing: those clocks can't quietly agree, because of everything in the last two phases.
GPS satellite (~20,000 km up, moving fast)
⌚ two relativistic effects fight each other:
SPECIAL relativity: satellite moves fast → its clock runs SLOWER
GENERAL relativity: satellite is high up, → its clock runs FASTER
in weaker gravity than us
the two don't cancel — net result: the orbiting clock
gains time relative to clocks on the ground, every single day
│
▼
uncorrected, the position error piles up FAST
The satellite is moving fast, so by special relativity its clock ticks slow. But it's also far from Earth, higher up in weaker gravity, so by general relativity (the gravitational time dilation from Phase 2) its clock ticks fast. These two effects don't cancel. The gravitational one wins, and the net result is that each satellite clock runs ahead of ground clocks by a small, steady amount every day.
Small sounds harmless. It isn't. GPS turns timing errors into distance errors at the speed of light, so a clock off by even a fraction of a millionth of a second becomes a position off by hundreds of meters. Left uncorrected, the error compounds and GPS positions would drift by something on the order of kilometers per day. Your navigation would be useless by lunchtime.
So the engineers build the correction in. The satellite clocks are deliberately set to tick at a rate that, once both relativistic effects are accounted for, comes out right as seen from the ground. Relativity isn't a footnote in the GPS design — it's a daily, operational requirement. The system would not work without it.
What just happened: GPS needs both relativities at once, and they pull in opposite directions. The fact that the system works — that your phone knows where you are — is a continuous, real-time experiment confirming Einstein, running in the sky right now.
Starlight that bends: gravitational lensing
If mass curves spacetime, then light passing a massive object should follow that curve — its path should bend, even though light has no mass to "pull." This was the first dramatic test of general relativity.
During a total solar eclipse, with the Sun's glare blocked, astronomers can see stars whose light grazed the edge of the Sun on its way to us. General relativity predicts those stars should appear shifted from their normal positions, because the Sun's gravity bent their light. In 1919, an expedition measured exactly that shift, and it matched Einstein's prediction rather than the older, smaller value Newton's gravity allowed. It made Einstein world-famous overnight.
true position of star
✦ .
` . light path bends near the Sun's mass
` .
( SUN )
. '
. '
. '
👁 you see the star shifted from where it "really" is
Today this is everyday astronomy, not a one-off. Whole galaxies act as lenses, bending the light of more distant galaxies behind them into arcs, rings, and multiple images. It's called gravitational lensing, and astronomers now use it as a tool — to weigh galaxy clusters, to map invisible dark matter by the way it bends light, and to magnify objects too far away to see otherwise. Curved spacetime stopped being a prediction and became an instrument.
Mercury's stubborn orbit
Here's a piece of evidence that was sitting in the data before Einstein, accusing Newton.
Mercury's orbit isn't a closed loop; the whole ellipse slowly rotates over time, its closest point to the Sun creeping around. Most of that creep, astronomers could explain with the tugs of the other planets. But a small leftover remained — a tiny, persistent drift that Newton's gravity could not account for, no matter how carefully they checked. For decades it was an unsolved scandal. Some proposed an unseen planet, "Vulcan," nudging Mercury. It was never found.
When Einstein applied general relativity to Mercury — the planet closest to the Sun, deepest in the Sun's curved spacetime — the equations produced extra orbital drift of precisely the missing amount. No new planet. No fudging. The leftover was the signature of curved spacetime, and it had been hiding in the observations the whole time, waiting for the right theory. For Einstein, this was the moment he knew he was right.
Gravitational waves: hearing spacetime ring
The boldest prediction took a century to confirm. If spacetime is a real, flexible fabric, then violent events should send ripples through it — waves of stretching and squeezing spacetime, spreading outward at the speed of light. Einstein predicted these gravitational waves in 1916, then doubted they'd ever be detectable, because they're almost unimaginably faint by the time they reach us.
In 2015, a pair of detectors called LIGO caught one. Two black holes, far across the universe, had spiraled into each other and merged, and the collision shook spacetime hard enough that the ripple, after traveling for over a billion years, still stretched and squeezed LIGO's kilometers-long arms by a distance far smaller than a single proton. The detectors measured it. The signal matched what general relativity predicted for two merging black holes, down to the shape of the final "ringdown." It earned a Nobel Prize, and it opened a brand-new way to observe the universe — not with light, but by feeling spacetime itself vibrate. Many more have been detected since.
two black holes spiral in and merge
◯ ◯ → ◯◯ → ● (one black hole)
\ /
\/ ripples in spacetime spread outward at light speed
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~> (over a billion years later...)
LIGO's arms stretch by less than
a proton's width — and we measured it
What just happened: A prediction Einstein himself doubted could ever be tested was confirmed a century later, by directly detecting spacetime flexing from a collision more than a billion light-years away. The fabric is real, and it rings.
Keeping it honest
So how settled is all this? The core of special and general relativity is about as solid as physics gets. It's been tested across an enormous range of scales — atomic clocks on towers and airplanes, particle accelerators, the orbits of planets and pulsars, the bending and lensing of light, and now gravitational waves — and it has passed every test, often to staggering precision. When you use GPS, you're trusting your life and your sense of direction to it.
What isn't finished is the very extreme: at the center of a black hole and at the first instant of the universe, general relativity's equations break down, predicting infinities that signal the theory is incomplete. Reconciling relativity with the quantum world (see /guides/what-physics-actually-is for where physics draws its current edges) is the great unsolved problem. But that's the frontier, not a crack in what you've learned here. Everything in this guide — time dilation, curved spacetime, the equivalence principle, the constancy of light — is confirmed, working, and woven into technology you use every day.
You started this guide being told relativity is only for geniuses. You now know its one rule, how it bends time and space, why gravity is geometry, and how we know it's true. That's not trivia. That's a working model of how space, time, and gravity actually behave — the same model a physicist carries, without the equations getting in the way of the picture.
[
{
"q": "Why does GPS require relativistic corrections to work?",
"choices": [
"Satellites are too heavy to track without it",
"Satellite clocks are affected by both special relativity (motion slows them) and general relativity (weaker gravity speeds them up); uncorrected, position errors would grow by kilometers per day",
"Relativity makes the radio signals travel faster",
"It only matters once per year"
],
"answer": 1,
"explain": "Both effects act on the orbiting clocks and don't cancel — the gravitational one wins, so the clocks run ahead. Since GPS converts timing into distance at light speed, uncorrected drift would reach kilometers per day."
},
{
"q": "What was special about Mercury's orbit that supported general relativity?",
"choices": [
"It was perfectly circular",
"It stopped moving entirely",
"Its orbit drifted by a small amount Newton's gravity couldn't explain, and general relativity predicted exactly that leftover",
"It was the same as every other planet's orbit"
],
"answer": 2,
"explain": "Mercury's orbital ellipse slowly rotates. A small leftover drift defied Newtonian gravity for decades. Being deepest in the Sun's curved spacetime, Mercury showed the extra drift general relativity predicts — matching precisely."
},
{
"q": "What did LIGO detect in 2015?",
"choices": [
"A new planet near the Sun",
"Gravitational waves — ripples in spacetime from two black holes merging over a billion light-years away",
"That the speed of light had changed",
"That gravity does not exist"
],
"answer": 1,
"explain": "LIGO directly detected gravitational waves: ripples in spacetime from merging black holes, stretching its kilometers-long arms by less than a proton's width. It confirmed a prediction Einstein doubted could ever be tested."
}
]
← Phase 2: General Relativity: Gravity Is Curved Spacetime · Overview
Check your understanding 3 questions
1. Why does GPS require relativistic corrections to work?
2. What was special about Mercury's orbit that supported general relativity?
3. What did LIGO detect in 2015?