Why Is Simultaneity Relative?

Short answer

Two events that happen at different places can be simultaneous in one inertial frame and not simultaneous in another moving relative to it. The reason is not that one observer receives the light from an event later. Each frame can account for signal travel time and still assign different coordinate times to the events.

To compare the times of distant events, a reference frame needs clocks at different locations and a defined synchronization procedure. Different inertial frames do not share one universal network of synchronized clocks. Their space and time coordinates are related by Lorentz transformations, which mix space and time while preserving the same invariant vacuum speed c.

So a pair of separated events can have Δt = 0 in one frame but Δt′ ≠ 0 in another. This frame-dependence applies to spacelike-separated events, not to events that can causally influence each other.

What Does “At the Same Time” Actually Mean?

For two events at the same place, simultaneity is local. If two sparks occur at one location and one moment, every inertial observer agrees that the events coincide.

The strange part begins when events are far apart.

Suppose one flash occurs at the left end of a long platform and another at the right end. What does it mean to say they happened “at the same time”?

One wristwatch cannot answer that question. A clock beside the left flash tells you the time at the left end, not the time at the right end. To assign times to events spread across space, you need clocks at different places and a rule for synchronizing them.

Newtonian physics treats distant simultaneity as universal. Special relativity does not.

That is the real source of the relativity of simultaneity.

Why Distant Clocks Need to Be Synchronized

A reference frame can be pictured as a grid of positions with clocks placed throughout it. To say that a distant event occurred at 3:00, those clocks need to share a synchronization procedure.

A standard procedure is Einstein synchronization. Take two clocks, A and B, at rest in the same inertial frame. A sends a light signal to B, B reflects it, and the signal returns to A. If the departure time at A is t₁ and the return time is t₂, the clock at B is assigned the midpoint time tB = (t₁ + t₂) / 2 when the signal reaches B. This procedure defines the frame's coordinate time by taking the light's one-way travel time to be equal in the two directions within that frame.

The procedure is not a claim that distant clocks are physically touching or that a signal arrives everywhere at once. It is a way to extend a local time standard across one inertial frame.

Once that network is synchronized, the frame can assign times to distant events. A second inertial frame moving relative to the first has its own synchronized clocks, and the two networks do not generally agree on which widely separated events share the same time coordinate.

The Train-and-Platform Example

Consider a long train moving to the right past a platform.

Two lightning strikes hit the track: one near the rear of the train and one near the front. Suppose the platform frame assigns both strikes the same time. According to clocks synchronized along the platform, the strikes are simultaneous in that frame.

Now use the train's frame. The train has its own clocks, synchronized with one another according to the same type of procedure carried out in the train frame. Those clocks do not generally assign the same time to the two strike events.

For the moving train frame, one strike receives an earlier coordinate time than the other.

The important point is not which flash reaches a passenger first. If a passenger sees one flash first merely because that light travelled a shorter distance, that is ordinary signal delay. It can be calculated and accounted for.

Relativity of simultaneity remains after each frame accounts for signal travel time using its own synchronized clock network. The disagreement is about the coordinate times assigned to the strike events themselves, not about which flash entered someone's eyes first.

Why This Is Not Just Light-Travel Delay

Seeing firework A before firework B does not prove A happened first; A may simply be closer. Physics can account for that propagation time.

The two questions are different:

When did the signal reach me?
What coordinate time does my frame assign to the event after accounting for propagation?

Relativity of simultaneity concerns the second question. Two inertial frames can account for signal travel time and still assign different coordinate times, so the effect is not an optical illusion.

How Two Frames Assign Different Times

The reason is tied directly to the invariant speed of light.

If all inertial observers measure the same vacuum light speed, their coordinates cannot be related by the old Galilean rule in which everyone shares one absolute time.

Instead, inertial frames are related by Lorentz transformations. Changing frames changes both the space coordinate and the time coordinate assigned to an event.

A useful picture is that each frame has its own “slice of now” across distant locations. Event L and Event R can share one time coordinate in Frame A but receive different time coordinates in moving Frame B. Neither frame is ignoring signal travel time; each uses its own internally consistent spacetime coordinates.

This is directly connected to why the speed of light is the same for every inertial observer. The Lorentz structure that preserves c is also the structure that removes universal distant simultaneity.

The Simple Lorentz Explanation

For two events separated by a time Δt and a distance Δx in one inertial frame, another frame moving at velocity v along the same direction assigns:

Δt′ = γ(Δt − vΔx/c²)

where:

γ = 1 / √(1 − v²/c²)

Suppose the events are simultaneous in the first frame:

Δt = 0

If they happen at different places, then Δx ≠ 0. The transformed time difference becomes:

Δt′ = −γvΔx/c²

For a genuinely moving second frame, that is generally not zero.

So:

simultaneous in one inertial frame does not generally mean simultaneous in another.

This is not an extra rule pasted onto relativity. It follows directly from the same transformation that relates space and time between inertial frames.

Which Event Orders Can Change?

Not every pair of events can swap order. The decisive question is the spacetime interval between them: could a signal travelling at or below the speed of light connect them?

Timelike-separated events

There is enough time between the events for an object moving slower than light to travel from one to the other. If Event A can cause Event B through an ordinary sub-light-speed process, all inertial observers agree that A comes before B.

Lightlike-separated events

One event can be connected to the other by a light signal. Their causal order is also preserved for all inertial observers.

Spacelike-separated events

The events are too far apart, relative to their time separation, for even light to travel from one to the other. These are the event pairs whose order can depend on the inertial frame.

One frame can say A happened first, another B first, and a third can make them simultaneous. No causal contradiction follows because spacelike-separated events cannot influence one another with a signal travelling at or below c.

Why Causality Is Still Safe

If Event A can physically cause Event B with a signal travelling at or below c, their separation is timelike or lightlike, and every inertial observer preserves their order. No inertial Lorentz transformation places a genuine effect before its cause.

Only spacelike-separated events can reverse coordinate order. The light-cone structure therefore protects causality: some distant time orderings are frame-dependent, but causal order is not.

How This Connects to Time Dilation

Relativity of simultaneity and time dilation are parts of the same Lorentz structure.

Two inertial observers can each say that the other's moving clocks run slow. The comparison involves clocks at different locations, and each frame uses its own definition of which distant readings count as simultaneous. Because those simultaneity slices differ, the two observers are not making the same set of distant clock comparisons.

For the full light-clock explanation and what “moving clocks run slow” actually means, see why time dilation happens.

Common Misconceptions

“It is only because light takes time to reach the observer.”

No. Signal delay can be accounted for. Frame-dependent simultaneity remains after that accounting.

“One observer is right; the other just sees the event late.”

No. Two inertial frames can use valid synchronized clock networks and still assign different coordinate times to separated events.

“If event order can change, causality must break.”

No. Only spacelike-separated events can reverse order. Timelike and lightlike causal order is preserved.

“Any two events can swap order.”

No. The spacetime separation between the events determines whether this is possible.

“Relativity of simultaneity is a separate weird rule.”

No. It follows from the same Lorentz structure behind invariant c, time dilation, and length contraction.

“There must be one hidden true universal now.”

Standard special relativity does not provide a preferred inertial frame with a universal distant simultaneity relation. Whether one wants to add further metaphysical structure is a different question and is not needed to explain the physics here.

Visual Explanation

Same events, different slices of “now”

Frame A’s synchronized clocks assign the distant events the same coordinate time. Frame B’s own Einstein-synchronized clock network assigns them different coordinate times.

The difference is not when light arrives at an observer. Only spacelike-separated events can have frame-dependent order; timelike and lightlike causal order is preserved.

One Thing to Remember

Relativity of simultaneity is not about observers being fooled by delayed light.

It means that different inertial frames do not share one universal network of synchronized distant clocks.

Because Lorentz transformations mix space and time, two separated events that share one time coordinate in one frame can receive different time coordinates in another.

But there is a strict limit: the order of causally connected events remains protected.