Length Contraction
A metre stick is a metre long. Surely that is one thing everyone can agree on — hand it to anyone,
anywhere, and they measure the same metre. But time dilation already taught us not to
trust "obvious" facts about space and time. If a moving clock runs slow, what happens to a moving
ruler? The answer is just as startling: a moving object is shortened along its
direction of motion. The faster it goes, the shorter it gets — not because it is squeezed or
stressed, but because length itself is measured differently in a frame the object is racing through.
This is length contraction, the spatial partner of time dilation, and it is
governed by exactly the same Lorentz factor \gamma. Where time stretches,
space shrinks — two sides of one coin. On this page we pin down the formula, meet the "proper
length", watch a rod contract as you crank up its speed, and confront the delicious barn-and-pole
puzzle that seems to break logic until relativity quietly rescues it.
Proper length, and the shrink factor
Start with a clean definition. The proper length L_0 of
an object is its length measured in the frame where it is at rest — the length you would get
by laying a tape measure alongside it at your leisure. A rocket's proper length is what the engineers
who built it measured in the hangar. Now let that rocket fly past you at speed
v. The length L you measure is
-
The contraction formula. An object of proper length
L_0 moving at speed v is measured to be
L = \dfrac{L_0}{\gamma} = L_0\sqrt{1 - v^2/c^2} along the direction of
motion, where \gamma = \dfrac{1}{\sqrt{1 - v^2/c^2}} \ge 1.
-
Only along the motion. Distances perpendicular to the velocity are
unchanged — a moving sphere keeps its width and height but is flattened front-to-back into an
ellipsoid.
-
Proper length is the longest. Because \gamma \ge 1,
everyone who sees the object moving measures it shorter than its rest length. The rest
frame wins the tape-measure contest.
Notice the symmetry with time dilation. There, a moving clock's interval is
multiplied by \gamma (time gets longer); here, a moving object's
length is divided by \gamma (space gets shorter). Same factor,
opposite direction. That is not a coincidence — the next lesson, the Lorentz transformation, shows
both falling out of one set of equations.
Watch a rod contract
Drag the speed slider from a standstill up towards the speed of light. The top bar is the rod at
rest — its full proper length L_0, fixed for reference. The bar below is
the same rod as measured in a frame it flies through at
\beta = v/c: it shrinks by the factor
1/\gamma = \sqrt{1-\beta^2}. Near \beta = 0 the
two are almost equal; push toward \beta = 1 and the moving rod collapses
toward a point.
Worked examples
Example 1 — a passing rocket. A rocket has proper length
L_0 = 100\ \text{m} and flies past Earth at
\beta = 0.8, so \gamma = 1/\sqrt{1-0.64} = 1.667.
Its measured length in the Earth frame is
L = \frac{L_0}{\gamma} = \frac{100\ \text{m}}{1.667} = 100\sqrt{1-0.64} = 100\times 0.6 = 60\ \text{m}.
The Earth measures the rocket to be just 60\ \text{m} long. The
astronauts inside, at rest with respect to their ship, still measure the full
100\ \text{m} — and they measure Earth to be squashed instead.
Example 2 — the muon's point of view. Recall the atmospheric muon. From the
ground, it survives because its clock runs slow (time dilation) and so lives long enough to
cross 15\ \text{km}. But how does the muon itself, in whose frame its
lifetime is only 2.2\ \mu\text{s}, make the trip? Because in the muon's
frame it is the atmosphere that is rushing past, and the
15\ \text{km} depth is contracted. At
\gamma \approx 22.4,
L = \frac{15\ \text{km}}{22.4} \approx 0.67\ \text{km} = 670\ \text{m}.
Only 670 metres of contracted atmosphere to cross, easily done in
2.2\ \mu\text{s} at nearly c. Two
frames, two explanations, one outcome — the muon reaches the ground. The ground calls it
time dilation; the muon calls it length contraction. They must agree on the physical fact, and they
do.
Example 3 — finding the speed. A spaceship's proper length is
200\ \text{m}, but a station measures it as 120\ \text{m}
as it flashes by. How fast is it going? The contraction factor is
L/L_0 = 120/200 = 0.6 = \sqrt{1-\beta^2}, so
1 - \beta^2 = 0.36 \;\Longrightarrow\; \beta^2 = 0.64 \;\Longrightarrow\; \beta = 0.8.
It is travelling at 0.8c.
The barn-and-pole "paradox"
Here is the puzzle that makes length contraction feel like a magic trick. A runner carries a pole of
proper length 10\ \text{m} toward a barn that is only
8\ \text{m} long, open at both ends. The farmer, standing by the barn,
reasons: "If the runner sprints fast enough, the pole contracts below 8\ \text{m},
so for one instant the whole pole fits inside my barn — and I can slam both doors shut on
it." At \gamma = 1.25 (that is \beta = 0.6) the
pole shrinks to 10/1.25 = 8\ \text{m}; a touch faster and it fits with room
to spare.
But now ride with the runner. In the pole's frame it is the barn that is rushing
toward the pole and contracting — the barn is now shorter than 8\ \text{m},
while the pole keeps its full 10\ \text{m}. There is no way the
pole ever fits inside. So does it fit or not? Both observers cannot be right… can they?
They can, and the rescue is the relativity of simultaneity.
"The pole fits" means "both doors are shut at the same time with the pole inside." But
simultaneity is frame-dependent! In the farmer's frame the two doors do close at the same instant,
briefly trapping the contracted pole. In the runner's frame the doors close at different
times — the far door opens and shuts to let the pole's tip through, then the near door closes behind
its tail, and the pole is never fully enclosed. No contradiction: they disagree about the pole
fitting only because they disagree about "at the same time." Everyone agrees on what physically
happens — no collision, the pole passes through — they just narrate the door timings differently.
Watch out — length contraction is a real measurement, not an illusion of your eyes,
but it is also easy to overclaim in the other direction. "Real" here means: if you set up two
synchronized markers in your frame and record the positions of the rod's two ends
simultaneously (in your frame), those positions really are only L_0/\gamma
apart. Nothing is compressed, no atoms are crushed; the object is under no stress in its own frame.
It is space measurement itself that differs between frames.
And crucially, the effect is reciprocal, exactly like time dilation. If you see my
rod contracted, I see yours contracted by the same factor — because each of us measures the
other's ends "at the same time" using our own, disagreeing notion of simultaneity. Neither rod is
"really" the short one. The symmetry only breaks, as always, if someone accelerates. Do not fall for
"but which rod is truly contracted?" — the honest answer is each, as seen by the other.
Surprisingly, no — and this fooled physicists for fifty years. Length contraction is what
you get when you record both ends of the object simultaneously in your frame. But a
camera does not do that: it captures light that all arrives at the lens at one instant, having left
different parts of the object at different earlier times, because the far side is further
away. When you work through those light-travel delays (James Terrell and Roger Penrose did, in 1959),
the two effects combine so that a fast-moving object does not appear flattened at all — it appears
rotated, as if you were seeing its side. A speeding sphere still photographs as a
perfect circle. So the measured contraction is real, but the naive "it would look squashed in a
snapshot" is wrong. Measuring and photographing are different acts — a lovely reminder to say
precisely what you mean by "see."