Time Dilation

Two identical clocks are synchronized and then separated. One stays on Earth. The other boards a spacecraft, travels fast, and returns. When the two clocks are placed side by side again, they no longer agree. The travelling clock shows less elapsed time, and so does every biological and chemical process that travelled with it.

This is not a malfunction and not an illusion. It is one of the best-tested predictions in physics. Particle accelerators, flying atomic clocks, satellites, and laboratory clocks separated by a few centimetres all show it. The effect is called time dilation, and it follows from two ideas that seem harmless on their own: the laws of physics are the same for every observer moving at constant velocity, and every such observer measures the same speed of light.

The puzzle is not whether time dilation happens. The puzzle is how it can be symmetric. If each observer sees the other's clock running slow, how can one of them end up younger? Answering that question requires one more idea, the relativity of simultaneity, and it is the heart of this article.

A Clock Along a Path

Motion is always relative to something

A passenger reading on a train is at rest relative to the train and moving at 100 km/h relative to the tracks. Neither description is the true one. Velocity only has meaning relative to a chosen reference frame, meaning a set of rulers and synchronized clocks that an observer uses to label where and when things happen.

Galileo already understood this for mechanics. In the nineteenth century it created a problem, because Maxwell's theory of electromagnetism predicted a single speed for light, $c \approx 299\,792\,458$ m/s, without saying relative to what (see Maxwell's Equations). Einstein's 1905 answer was to accept both claims at face value: no inertial frame is preferred, and light moves at $c$ in all of them. ^[1]

Why a moving clock must tick slower

Imagine a clock made of two mirrors facing each other, with a pulse of light bouncing between them. Each round trip is one "tick".

For an observer riding with the clock, the light moves straight up and down. For an observer watching the clock fly past, the light must follow a longer diagonal zigzag, because the mirrors move sideways while the pulse is in flight. Both observers agree the light travels at the same speed $c$. A longer path at the same speed takes more time. So the stationary observer concludes that each tick of the moving clock lasts longer than a tick of their own identical clock.

Nothing is special about light clocks. If a moving light clock ran slow while a moving mechanical clock beside it did not, a traveller could detect their "absolute" motion by comparing them, which would violate the first principle. All physical processes in the moving system, including atomic vibrations, radioactive decay, and heartbeats, must slow by the same factor.

The Formal Description

The Lorentz factor

Working through the geometry of the light clock gives the central quantity of special relativity, the Lorentz factor:

Here:

• $v$ is the relative speed between the clock and the observer;
• $c$ is the speed of light;
• $\Delta\tau$ is the proper time, the time measured by the clock itself;
• $\Delta t$ is the time between the same two ticks, measured in the frame where the clock moves.

Because $\gamma \ge 1$, the coordinate time $\Delta t$ is always at least as long as the proper time. At everyday speeds, $\gamma$ is almost exactly 1. For a jet at 250 m/s, $\gamma - 1 \approx 3.5 \times 10^{-13}$. The factor only becomes large as $v$ approaches $c$:

Proper time as a length in spacetime

In 1908 Hermann Minkowski showed that special relativity has a natural geometric form: space and time combine into a four-dimensional spacetime, and the quantity every observer agrees on is not the time interval or the distance separately, but the combination

A clock following a path through spacetime (its worldline) records the proper time

This formula is the cleanest way to think about ageing in relativity. Each clock measures the "length" of its own worldline. Two worldlines that begin and end at the same events can have different lengths, just as two roads between the same towns can have different lengths. In the geometry of spacetime, the straight path (unaccelerated, inertial motion) is the one with the longest proper time. That minus sign in the formula is where the familiar intuition from Euclidean geometry is reversed.

Length contraction and the relativity of simultaneity

The same principles that dilate time also change distances and "now". The full relation between two inertial frames moving at speed $v$ along $x$ is the Lorentz transformation:

Two consequences follow:

• Length contraction. An object of length $L_0$ in its own rest frame measures $L = L_0/\gamma$ along the direction of motion in a frame where it moves.
• Relativity of simultaneity. The term $v x/c^2$ means that two events at different places that are simultaneous in one frame ($\Delta t = 0$) are generally not simultaneous in another ($\Delta t' = -\gamma v\,\Delta x/c^2$). The further apart the events, the larger the disagreement.

The second point is the one people usually skip. It is also the one that dissolves the paradoxes.

The Symmetry Puzzle

Suppose a spacecraft cruises away from Earth at constant speed. Earth's physicists, using Earth's frame, find that the spacecraft clock ticks slowly. The traveller, using the spacecraft's frame, is equally entitled to say that Earth is moving, and finds that Earth's clock ticks slowly.

Both statements are correct, and they do not contradict each other. To compare two clocks that are far apart, each observer must decide which reading on the distant clock is happening "at the same time" as a reading on the nearby clock. Earth and the traveller use different definitions of "at the same time", because their frames disagree about simultaneity. Each observer compares the other's clock against their own set of simultaneous events, and each correctly finds the other's clock behind.

An analogy from ordinary geometry helps. Two surveyors each lay a metre stick along their own "forward" direction, but their forward directions differ by some angle. Each measures the other's stick by projecting it onto their own forward axis, and each truthfully finds the other's stick shorter than a metre. There is no contradiction, because "along my axis" means something different for each surveyor. In relativity, "at the same time" for distant events means something different for each frame, and mutual time dilation is the spacetime version of that mutual shortening.

While both observers move inertially and stay apart, the question "whose clock is really slower?" has no frame-independent answer. It only acquires a definite answer when the clocks are brought back together.

The Twin Paradox and the Simultaneity Jump

The asymmetry is the turnaround

Now let the traveller return. Earth stays in one inertial frame the entire time. The traveller does not: to come home, they must decelerate, turn around, and accelerate back. For the outbound leg they belong to one inertial frame; for the return leg, to a different one.

That change of frame is what breaks the symmetry. The two worldlines connect the same departure event and the same reunion event, but only Earth's is straight. By the proper-time formula, the straight worldline is the longer one. The traveller is younger at the reunion, and every observer, in every frame, agrees on that final comparison, because it is a comparison of two clocks at the same place and time.

The acceleration itself is not what causes the age difference. One can design versions of the problem in which both twins undergo identical accelerations and still age differently, because their paths differ in length. The acceleration matters because it marks which twin changed frames.

What the traveller calls "now" on Earth

The change of frame has a striking consequence. Before the turnaround, the traveller's "line of simultaneity" (the set of events they consider to be happening now) intersects Earth's history at one moment. After the turnaround, in the new frame, the line of simultaneity is tilted the other way and intersects Earth's history at a much later moment. If the traveller insists on labelling distant events with the "now" of whichever inertial frame they currently occupy, Earth's clock appears to leap forward during the turnaround. This is often called the simultaneity jump.

This must be read carefully:

• Established fact: the traveller comes back younger by exactly the amount the proper-time formula predicts.
• Coordinate convention: the "jump" is a change in which distant Earth event the traveller labels as simultaneous with their own present. It is a bookkeeping change caused by switching frames, not a physical event on Earth.
• What does not happen: Earth does not skip time. Nobody on Earth experiences a sudden lurch, and Earth's history is continuous in its own frame.

A good test of this reading is to ask what the traveller actually sees through a telescope. Light from Earth arrives continuously. On the outbound leg, the Earth images are strongly redshifted and Earth's clock appears to run very slowly; on the return leg, the images are blueshifted and Earth's clock appears to run very fast. There is no discontinuity in what is observed, only a change in the rate at which the images arrive. The "jump" appears only when one converts those observations into a claim about what is happening "now" far away.

A Worked Example: A Planet 100 Light-Years Away

A trip to a star 100 light-years from Earth need not take a human lifetime for the traveller. Take a ship that cruises at $v = 0.99c$ and ignore, for simplicity, the time spent accelerating.

Lorentz factor:

Earth's account. The distance is 100 light-years in Earth's frame, so the one-way trip lasts

Traveller's account. The traveller's clock records

The traveller does not feel time passing slowly. Their clock, their thoughts and their metabolism run normally. From the traveller's point of view, the explanation is length contraction: in the ship's frame, the Earth–planet distance is not 100 light-years but $100/7.09 \approx 14.1$ light-years, and covering that at $0.99c$ takes about 14.2 years. Earth explains the result by time dilation; the traveller explains it by length contraction. The two accounts describe the same physics from different frames and agree on every reading of every clock.

The round trip and the jump, in numbers. If the traveller immediately turns around and comes home at the same speed, Earth ages about $2 \times 101 \approx 202$ years and the traveller about $2 \times 14.2 \approx 28.5$ years.

The simultaneity jump can also be computed. Just before the turnaround, the Lorentz transformation says the traveller's frame regards Earth's clock as reading about $101 - 0.99 \times 100 \approx 2$ years. Just after switching to the return frame, it regards Earth's clock as reading about $101 + 99 \approx 200$ years. From the traveller's changing perspective, Earth ages about 2 years on the way out, about 198 years "during" the turnaround, and about 2 years on the way back. The total, 202 years, agrees with Earth's own records.

Meanwhile, what the traveller sees at the moment of turnaround is light that left Earth when Earth's clock read only about 1 year. On the way home, they watch roughly 201 years of Earth's history unfold, fast-forwarded by a Doppler factor of about 14, over their 14.2-year return. The observations are smooth. Only the coordinate labels jump.

These numbers are a mathematically exact consequence of special relativity. Actually building a ship that reaches $0.99c$ is another matter: the kinetic energy required per kilogram of ship is about $(\gamma - 1)mc^2 \approx 5.5 \times 10^{17}$ joules, more than the energy released by a hundred megatons of TNT. Interstellar dust, radiation, and propulsion are unsolved engineering problems.

Gravitational Time Dilation

Special relativity concerns motion. General relativity adds a second, independent effect: clocks deeper in a gravitational field run slower than clocks higher up. Near Earth's surface, the fractional difference between two clocks separated by a height $h$ is approximately

where $g \approx 9.8$ m/s² is the local gravitational acceleration. For one metre, this is about $1.1 \times 10^{-16}$, or roughly 3 nanoseconds per year. Near a black hole the effect becomes enormous.

The effect was first measured as a shift in the frequency of gamma rays sent up a 22.5-metre tower at Harvard by Pound and Rebka in 1960. ^[11] In 1976, the Gravity Probe A rocket carried a hydrogen maser clock to about 10,000 km altitude and confirmed the prediction to roughly one part in ten thousand. ^[12] In real situations, such as satellites, both effects act at once and must be added.

Historical Development

Before Einstein, Lorentz, FitzGerald and Poincaré had already written down transformations resembling those above, as mathematical devices to explain why experiments failed to detect Earth's motion through a hypothetical "ether". Lorentz even introduced a "local time" variable. What they lacked was the interpretation. Einstein's 1905 paper treated the transformations as describing real properties of space and time, derived them from two principles, and noted that a clock moved around a closed path and returned would lag behind a clock that stayed put. ^[1] Minkowski's 1908 geometric reformulation turned this into the picture of worldlines with different lengths.

The twin paradox was raised almost immediately, often as an objection to the theory. The objection rested on assuming that the situation is symmetric. It is not, and the resolution has been standard physics for about a century.

Evidence

Time dilation has been measured in several completely different ways, with independent clocks, over speeds ranging from a sprinter's pace to more than $0.999c$.

Cosmic-ray muons

Muons are unstable particles created when cosmic rays strike the upper atmosphere, about 10–15 km up. At rest, a muon's mean lifetime is about 2.2 microseconds. Even at nearly light speed, a particle living 2.2 μs would travel only about 660 m on average before decaying. Without time dilation, very few muons would reach the ground. Many do.

In 1941, Rossi and Hall compared the muon flux at two altitudes in Colorado and found that the survival rate depended on muon momentum in the way time dilation predicts: faster muons lived longer. ^[2]

In 1963, Frisch and Smith made the effect quantitative. They selected muons with speeds between about $0.9950c$ and $0.9954c$, counted them on top of Mount Washington in New Hampshire (about 1,900 m), and then counted the same population at sea level in Cambridge, Massachusetts. They measured about 563 per hour at the summit and about 408 per hour at sea level. Without time dilation, only a few dozen per hour should have survived the descent. The inferred dilation factor, $8.8 \pm 0.8$, agreed with the predicted value. ^[3]

The CERN muon storage ring

In 1977, a CERN team stored muons in a circular ring at $\gamma \approx 29.3$ and measured their decay. The muons lived about 64.4 μs instead of 2.2 μs, in agreement with the Lorentz factor to about two parts in a thousand. ^[4] This experiment also tested whether the enormous acceleration of circular motion (around $10^{18}$ times Earth's gravity) affects the clock rate beyond what speed alone predicts. It did not, to the precision of the measurement, which supports the view that proper time depends on the path, not on acceleration directly.

Flying atomic clocks

In October 1971, Hafele and Keating flew caesium-beam atomic clocks around the world on commercial airliners, once eastward and once westward, and compared them with reference clocks at the U.S. Naval Observatory. Because Earth rotates, the eastward clocks moved faster relative to a non-rotating frame than the ground clocks, and the westward ones moved slower; both flights also sat higher in Earth's gravitational field. Theory predicted a loss of about 40 ns eastward and a gain of about 275 ns westward. The clocks lost about 59 ns and gained about 273 ns, consistent within the stated uncertainties. ^[5,6] The experiment was relatively crude by modern standards, but it was a direct demonstration with macroscopic clocks.

GPS

The Global Positioning System is an everyday engineering application of both kinds of time dilation. GPS satellites orbit at about 20,200 km altitude and about 3.9 km/s. Their speed makes their clocks run slower than ground clocks by about 7 microseconds per day; their height makes them run faster by about 45 microseconds per day. The net effect is about 38 microseconds per day fast. ^[7]

That sounds small. But GPS positions come from light-travel times, and 38 microseconds of light travel corresponds to over 11 km. The system compensates by tuning the satellite clocks slightly low before launch (a fractional offset of about $4.465 \times 10^{-10}$) and by further corrections in receiver software. ^[7] Without relativity, uncorrected timing errors would quickly make the system useless for navigation.

Optical clocks at walking speed

The most personal-scale test came in 2010. Chou and colleagues at NIST compared two aluminium-ion optical clocks, among the most precise clocks ever built. When one ion was set moving at speeds below 10 metres per second, roughly a sprinter's pace, its clock rate shifted as special relativity predicted. When one clock was raised by only 33 cm, it ran faster by about 4 parts in $10^{17}$, matching gravitational time dilation. ^[8]

Spectroscopic tests

Time dilation also appears as the "transverse" or second-order Doppler shift of light from moving atoms. Ives and Stilwell first detected it in 1938 with fast hydrogen ions. ^[9] A 2014 version at the GSI storage ring in Darmstadt used lithium ions at $0.338c$ and confirmed the relativistic relation between speed and $\gamma$ to about 2 parts per billion. ^[10]

Limitations and Common Misunderstandings

Time dilation does not mean time "feels" slower. Every observer experiences their own time normally. The effect only appears when comparing clocks between frames, or clocks reunited after different paths.

It does not allow travel into the past. Time dilation lets a traveller reach Earth's far future quickly by their own clock. It provides no way back. Travel to the past would require exotic spacetime structures such as closed timelike curves, which general relativity allows mathematically but which no experiment has shown to exist.

The simultaneity jump is not a physical event. It depends on a particular convention for assigning "now" to distant places. Other conventions, such as the one based on radar signals, produce a smooth change instead of a jump. None of them changes any measured clock reading.

Acceleration is not the cause, but it is not irrelevant. The age difference comes from the geometry of the paths. Acceleration is simply what allows a path to bend.

Deeper interpretation remains debated. That observers disagree on simultaneity is an established part of the theory. What it implies about the nature of time, for example whether all moments are equally real, is a philosophical question, discussed in Block Universe. Whether time itself is fundamental or emerges from something deeper is an open problem in quantum gravity.

Time Becomes Local

Time dilation is where relativity stops being abstract. It is measured in cosmic rays, accelerators, airliners, satellites, and laboratory clocks, and it has to be engineered into navigation systems used by billions of people daily.

It also changes what "time" means. There is no single clock ticking for the whole universe. Each object carries its own proper time along its own path, and what counts as "now" far away depends on how one is moving. The twin paradox is not a flaw in relativity. It is an illustration that the question "what time is it over there?" has no answer until one specifies a frame.

Moving clocks and clocks deeper in gravity run slow by the amounts relativity predicts. The same equations allow a traveller to cross a hundred light-years in a decade or so of personal time. Engineering such a journey remains far beyond us, and the relativity of simultaneity still leaves a philosophical question behind: what, if anything, makes the present uniquely real?

References

[1] Einstein, A. (1905). “Zur Elektrodynamik bewegter Körper” (On the Electrodynamics of Moving Bodies). Annalen der Physik, 322(10), 891–921. https://doi.org/10.1002/andp.19053221004

[2] Rossi, B., & Hall, D. B. (1941). “Variation of the Rate of Decay of Mesotrons with Momentum.” Physical Review, 59(3), 223–228. https://doi.org/10.1103/PhysRev.59.223

[3] Frisch, D. H., & Smith, J. H. (1963). “Measurement of the Relativistic Time Dilation Using μ-Mesons.” American Journal of Physics, 31(5), 342–355. https://doi.org/10.1119/1.1969508

[4] Bailey, J., et al. (1977). “Measurements of relativistic time dilatation for positive and negative muons in a circular orbit.” Nature, 268, 301–305. https://doi.org/10.1038/268301a0

[5] Hafele, J. C., & Keating, R. E. (1972). “Around-the-World Atomic Clocks: Predicted Relativistic Time Gains.” Science, 177(4044), 166–168. https://doi.org/10.1126/science.177.4044.166

[6] Hafele, J. C., & Keating, R. E. (1972). “Around-the-World Atomic Clocks: Observed Relativistic Time Gains.” Science, 177(4044), 168–170. https://doi.org/10.1126/science.177.4044.168

[7] Ashby, N. (2003). “Relativity in the Global Positioning System.” Living Reviews in Relativity, 6, 1. https://doi.org/10.12942/lrr-2003-1

[8] Chou, C. W., Hume, D. B., Rosenband, T., & Wineland, D. J. (2010). “Optical Clocks and Relativity.” Science, 329(5999), 1630–1633. https://doi.org/10.1126/science.1192720

[9] Ives, H. E., & Stilwell, G. R. (1938). “An Experimental Study of the Rate of a Moving Atomic Clock.” Journal of the Optical Society of America, 28(7), 215–226. https://doi.org/10.1364/JOSA.28.000215

[10] Botermann, B., et al. (2014). “Test of Time Dilation Using Stored Li⁺ Ions as Clocks at Relativistic Speed.” Physical Review Letters, 113, 120405. https://doi.org/10.1103/PhysRevLett.113.120405

[11] Pound, R. V., & Rebka, G. A. (1960). “Apparent Weight of Photons.” Physical Review Letters, 4(7), 337–341. https://doi.org/10.1103/PhysRevLett.4.337

[12] Vessot, R. F. C., et al. (1980). “Test of Relativistic Gravitation with a Space-Borne Hydrogen Maser.” Physical Review Letters, 45(26), 2081–2084. https://doi.org/10.1103/PhysRevLett.45.2081