Doppler-Shift Resolution of the Twin Paradox

Give each twin a strobe that flashes once a year, on their birthday, and a telescope. Now nobody has to argue about whose clock "really" runs slow, because there is nothing left to argue about: a flash is either seen or it is not. Counting is not a convention. The traveller flies out at $0.6c$ , turns at a star $3$ light-years away, and comes home. She counts the flashes from Earth: ten. Earth counts the flashes from her: eight. Those are integers. Every observer in the universe agrees on them, because they are just tallies of events that happened at a particular eye, and no Lorentz transformation can turn ten into nine. Ten and eight are the answer. She lived eight years; Earth lived ten. And the entire asymmetry is now visible, in the open, with no simultaneity convention anywhere: the traveller spends half her trip watching a redshift and half watching a blueshift — four years of each. Earth does not. The light announcing the turnaround has three light-years to cross before Earth can see it, so Earth watches redshifted flashes for eight of its ten years and blueshifted flashes for only two. That $8{:}2$ against her $4{:}4$ is not a detail. It is the resolution.

Key takeaways

Signature $(+,-,-,-)$ : $\mathrm{d}s^2 = c^2\mathrm{d}t^2 - \mathrm{d}x^2$ , timelike $>0$ .
Bondi's $k$ -factor $k = \sqrt{(1+\beta)/(1-\beta)}$ : the factor by which a receding source's signal period is stretched. It is the Doppler factor of `aq-sr-longitudinal-doppler-effect`, recalled — and in rapidity it is simply $k = e^{\varphi}$ .
Two identities do all the work: $\;\boxed{\gamma = \tfrac12\!\left(k + 1/k\right)}\;$ and $\;\boxed{\beta = \dfrac{k^2-1}{k^2+1}}\;$ . The second lets you read the speed off a radar echo with no Lorentz transform at all.
The pulse count is an invariant integer. Every flash emitted is a flash received. Counting them proves $\tau_{\text{trav}} = T_{\text{home}}/\gamma$ using no simultaneity convention whatsoever — which is exactly why this resolution needs no turnaround gap.
The asymmetry, made concrete: Earth sees red for $(L/v)(1+\beta)$ and blue for only $(L/v)(1-\beta)$ . The traveller sees red and blue for equal halves.
Key idea: the twins do not disagree about time, they disagree about when the news arrives. Light delay is the whole of it — and the light delay is unequal because only one twin changed frames.

The core insight

Replace "how fast is her clock running?" — a frame-dependent question with a convention buried in it — with "how many flashes did you see?" — an integer. The Doppler factor $k$ converts one twin's proper time directly into the other's seen rate, so the count is a pure product of $k$ 's and proper times. Because every flash sent is eventually received, the two tallies must balance, and making them balance forces $\tau_{\text{trav}} = T_{\text{home}}/\gamma$ . No simultaneity, no turnaround gap, no acceleration analysis. Just arithmetic on what a telescope sees.

Where this lives today

The $k$ -calculus is not a party trick for the classroom, though Bondi built it as one. It is how deep-space navigation actually works. A spacecraft's radial velocity is measured by exactly the procedure of Example 2: a station transmits at a known frequency, the craft coherently transponds, and the returned signal is compared with the outgoing one. The measured ratio is $k^2$ — two applications of the Doppler factor, one on each leg — and identity (2) turns it into $\beta$ directly. The two-way ranging that fixes the craft's distance is the radar construction of Step 3, and its time-tagging is the radar time. NASA's Deep Space Network runs this at $S$ - and $X$ -band on every mission it tracks. The same identities run relativistic astronomy. A jet knot approaching us at $\beta$ has its emission blueshifted by $k$ and its variability timescale compressed by the same factor; a receding counter-jet is dimmed and slowed by $1/k$ . The ratio of approaching to receding brightness is a power of $k^2$ , which is why one-sided jets are the norm — the physics is developed in `aq-sr-relativistic-beaming`, but the factor being raised to a power is precisely the number this article calls $k$ . And the flash count itself is the honest picture of any long-baseline timing experiment. A pulsar in a binary orbit is a strobe on a moving platform; a spacecraft carrying an atomic clock past a planet is a twin. What such experiments measure is never "the rate of a distant clock" — that phrase hides a convention. What they measure is arrival times, and arrival times are integers on a counter.

Origins · 2 minDoppler shift and how this was discovered

Doppler shift

1842 — present · Prague · Cambridge · Kings College London

The shift itself is Christian Doppler's (1842); its relativistic form and its role as the engine of the twin resolution belong to a different tradition, and the portrait therefore belongs to the concept rather than to one person. The decisive move was Hermann Bondi's. In Relativity and Common Sense: A New Approach to Einstein (Doubleday Anchor, 1964), Bondi built the whole of special relativity on a single measured number — the ratio $k$ by which a receding source's signal period is stretched — and derived time dilation, length contraction, velocity composition and the twin paradox from it, in that order, using no Lorentz transformation and, crucially, no synchronisation of distant clocks. He called it the $k$ -calculus. His argument is Steps 1–4 of the first derivation above, and his showcase was exactly this article's subject: he took the twin paradox, which everyone treated as a puzzle about simultaneity, and solved it with a tally of light flashes. What makes the $k$ -calculus more than a pedagogical device is what it leaves out. Simultaneity at a distance is a convention; radar echoes are not. Bondi's route reaches the same answers while touching only quantities a single observer with a single clock can measure, and it is the reason the pulse-count resolution feels less like an argument and more like an audit. The relativistic Doppler factor $k = \sqrt{(1+\beta)/(1-\beta)} = e^{\varphi}$ — the boost's eigenvalue on a null ray — turns out to be the most operationally direct quantity in the whole theory.

History

1842

— Christian Doppler predicts the frequency shift of waves from a moving source, in Über das farbige Licht der Doppelsterne.

1905

— Einstein derives the relativistic Doppler formula in "Zur Elektrodynamik bewegter Körper", including the time-dilation factor that classical Doppler lacks.

1911

— Paul Langevin poses the travelling twin in Scientia X, 31-54, and blames the asymmetry on acceleration.

1911

— Max von Laue gives the geometric resolution, locating the asymmetry in the traveller's use of two inertial frames.

1938

— Ives and Stilwell confirm the second-order (time-dilation) piece of the relativistic Doppler shift with hydrogen canal rays.

1964

— Hermann Bondi's Relativity and Common Sense builds special relativity on the radar factor k, and resolves the twin paradox by counting light flashes.

1971

— Hafele and Keating fly caesium clocks around the world; the east-west asymmetry is the laboratory echo of the twins.

The physical picture

Set the trip: coasting speed $\beta = 0.6$ , turnaround star at $L = 3$ light-years, so $\gamma = 1.25$ . Earth logs $T = 2L/v = 10$ years; the traveller logs $T/\gamma = 8$ years. Those totals are owned by `aq-sr-twin-paradox` and are not re-derived here — the point of this article is that you can rediscover them from nothing but a strobe and a telescope. The engine of the argument is a single number. If a source recedes from you at $\beta$ and flashes once per second of its proper time, you receive the flashes once every $k$ seconds of your proper time, with

k = \sqrt{\frac{1+\beta}{1-\beta}} \;=\; e^{\varphi}, \qquad \varphi = \operatorname{artanh}\beta .

At $\beta = 0.6$ this is $k = \sqrt{1.6/0.4} = \sqrt{4} = \mathbf{2}$ , exactly. The whole of this article is arithmetic with the number $2$ . (The formula itself is the longitudinal Doppler factor, owned by `aq-sr-longitudinal-doppler-effect`; the exponential form $k = e^{\varphi}$ says that $k$ is the eigenvalue of the boost on a null ray — recalled from `aq-sr-lorentz-boosts-hyperbolic-rotations` — which is why collinear Doppler factors multiply while rapidities add.) Now watch what each twin actually sees.

The traveller flips from receding to approaching at the instant she turns, which is the midpoint of her own trip: $4$ years out, $4$ years back. She sees Earth's flashes at rate $1/k = \tfrac12$ per year for the first half and $k = 2$ per year for the second. Total: $4 \times \tfrac12 + 4 \times 2 = 2 + 8 = \mathbf{10}$ flashes.

Earth flips when the news of the turnaround arrives — and that news is a light signal that must cross three light-years. The turnaround happens at Earth-time $L/v = 5$ yr, and the flash announcing it lands at Earth at $5 + 3 = \mathbf{8}$ yr. So Earth sees redshift for eight of its ten years and blueshift for only two. Total: $8 \times \tfrac12 + 2 \times 2 = 4 + 4 = \mathbf{8}$ flashes. That is the entire resolution, and notice what it did not use. It never asked what time it "was on Earth" while the traveller was in flight. It never invoked a line of simultaneity, never mentioned the $6.4$ -year gap of `aq-sr-twin-paradox-asymmetry`, never analysed the acceleration. It counted photons. The asymmetry is not hidden in a convention — it is sitting in plain sight in the fact that light takes three years to get home, and that the traveller is the one who was there when the turnaround happened.

Figure

PUNCHLINE — the two ledgers, for β = 0.6, L = 3 ly, k = 2. The traveller's experience is SYMMETRIC: 4 years of redshift, 4 of blueshift, and she tallies 2 + 8 = 10 of Earth's birthdays. Earth's experience is LOPSIDED: the turnaround light needs 3 years to get home, so Earth sees redshift for (L/v)(1+β) = 8 years and blueshift for only (L/v)(1−β) = 2 years, and tallies 4 + 4 = 8 of hers. Ten against eight. Both integers, both invariant, both agreed by everyone. That 8:2 against 4:4 IS the paradox's answer

The core equations

The $k$ -factor is the only input, and it is recalled, not derived:

k \;=\; \sqrt{\frac{1+\beta}{1-\beta}} \;=\; e^{\varphi}, \qquad \frac{1}{k} = \sqrt{\frac{1-\beta}{1+\beta}} = e^{-\varphi} \qquad (1)

where: $\varphi = \operatorname{artanh}\beta$ is the rapidity. Physical reading: $k$ multiplies a receding source's signal period; $1/k$ multiplies an approaching source's. That the two are exact reciprocals is the statement that reversing the motion reverses the sign of $\varphi$ . At $\beta = 0.6$ , $\varphi = \ln 2$ and $k = 2$ exactly. From (1), two identities that this article owns and uses relentlessly:

\boxed{\;\gamma \;=\; \tfrac12\!\left(k + \frac{1}{k}\right)\;} \qquad\qquad \boxed{\;\beta \;=\; \frac{k^{2}-1}{k^{2}+1}\;} \qquad (2)

where: both follow from (1) by one line of algebra. Physical reading: $\gamma$ is the arithmetic mean of the two Doppler factors. That is not a curiosity — it is the pivot of the whole pulse count, because the traveller spends exactly half her time in each. And the second identity is a measurement recipe: bounce a radar pulse, read $k$ , get $\beta$ — no Lorentz transform, no clock synchronisation, nothing but a stopwatch at one place. At $k=2$ : $\gamma = \tfrac12(2 + \tfrac12) = 1.25$ and $\beta = 3/5$ . The invariant that makes the count binding:

N^{\text{sent}}_{\text{trav}} = N^{\text{recv}}_{\text{home}}, \qquad N^{\text{sent}}_{\text{home}} = N^{\text{recv}}_{\text{trav}} \qquad (3)

where: $N$ counts flashes over the whole trip. Physical reading: every flash emitted before the reunion is received before the reunion. No flash is lost, none is created, and the tally is an integer that no observer can dispute. This is the only conservation law the argument needs, and it is the reason the resolution requires no convention. And the timing of the news — the article's structural asymmetry:

t_{\text{red}} = \frac{L}{v}(1+\beta), \qquad t_{\text{blue}} = \frac{L}{v}(1-\beta), \qquad t_{\text{red}} + t_{\text{blue}} = \frac{2L}{v} = T \qquad (4)

where: these are Earth-frame durations. Physical reading: Earth's redshift phase runs until the turnaround light arrives, at $L/v + L/c$ . At $\beta = 0.6$ , $L = 3$ ly: $t_{\text{red}} = 5(1.6) = 8$ yr and $t_{\text{blue}} = 5(0.4) = 2$ yr. The traveller's split is $4{:}4$ ; Earth's is $8{:}2$ . Nothing else in this article is doing any work.

Counting flashes

Derivation ABondi's radar: the k-factor and the speed, with no Lorentz transform

Before counting anything we should be clear that $k$ is not smuggled in from the Lorentz transformation. Bondi's insight was that $k$ can be defined operationally — as the number a radar set measures — and that special relativity's kinematics can then be reconstructed from it. This derivation does that: it takes only the principle of relativity and the constancy of $c$ , defines $k$ by a radar echo, and produces $\beta$ . It is the article's owned machinery and it is the reason the pulse count is trustworthy.

Step 1. Define $k$ operationally. Earth flashes twice, separated by $t_1$ of Earth's proper time. The receding traveller receives the two flashes separated by $k\,t_1$ of her proper time. The number $k$ depends only on the relative speed — by the principle of relativity it cannot depend on who is "really" moving, and by the constancy of $c$ it cannot depend on how far apart they are. Why: $k$ is defined by a measurement, not by a coordinate transformation. Nothing has been assumed about simultaneity or about how either party synchronises distant clocks — there are no distant clocks.

Step 2. Bounce the pulse. Earth sends a single radar pulse at Earth-time $t_1$ . The traveller receives it at her proper time $k\,t_1$ and immediately reflects it. Now she is the source and Earth is the receiver, still receding at $\beta$ , so by the same $k$ — this is where the principle of relativity earns its keep — Earth receives the echo at Earth-time $t_2 = k\,(k\,t_1) = \boxed{k^{2}\,t_1}$ . Why: the factor applies twice, once each way, because the relative speed is the same in both directions. Everything now follows from the single measured ratio $t_2/t_1 = k^2$ .

Step 3. Locate the reflection event by radar. Earth assigns the reflection a distance $d = \tfrac{c}{2}(t_2 - t_1)$ — half the round-trip light travel — and a time $t = \tfrac12(t_2 + t_1)$ , the midpoint of send and receive. Why: these are the definitions any radar operator uses, and they follow from $c$ being the same both ways. Note that this is the one place a convention enters, and it enters as the ordinary radar convention, not as a claim about relativity.

Step 4. Read off the speed. The traveller left Earth at $t = 0$ , so her speed is

\beta = \frac{d}{c\,t} = \frac{\tfrac12(t_2-t_1)}{\tfrac12(t_2+t_1)} = \frac{k^{2}t_1 - t_1}{k^{2}t_1 + t_1} = \boxed{\frac{k^{2}-1}{k^{2}+1}} .

Why: this is identity (2), derived from a radar echo and nothing else. Speed has been measured with one clock at one place.

Step 5. Invert and check against the Doppler formula. Solving for $k$ : $k^2 = (1+\beta)/(1-\beta)$ , so $k = \sqrt{(1+\beta)/(1-\beta)}$ — precisely the longitudinal Doppler factor of `aq-sr-longitudinal-doppler-effect`, arrived at from the opposite direction. And $\tfrac12(k + 1/k) = \tfrac12\big[(1+\beta) + (1-\beta)\big]/\sqrt{(1+\beta)(1-\beta)} = 1/\sqrt{1-\beta^2} = \gamma$ . Why: the two routes agree, as they must. But the radar route never wrote down a Lorentz transformation, which is the entire pedagogical point of the $k$ -calculus.

Step 6. Run the anchor. Earth pings at $t_1 = 2$ yr. With $k = 2$ , the echo returns at $t_2 = k^2 t_1 = 8$ yr. Radar distance: $\tfrac12(8-2) = 3$ ly. Radar time: $\tfrac12(8+2) = 5$ yr. Speed: $3/5 = 0.6$ . ✓ And the reflection event is $(5\ \text{yr},\ 3\ \text{ly})$ — exactly the turnaround. The traveller received that ping at her proper time $k t_1 = 4$ yr — exactly her own midpoint. Why: the radar construction lands on the turnaround event from both ends, which is not a coincidence: the ping that bounces off the turnaround is the same light-crossing that makes Earth's redshift phase last $8$ years. *Verified in verify_aq-sr-twin-paradox-doppler-resolution.py: $t_2 = k^2 t_1$ ; $\beta = (k^2-1)/(k^2+1)$ symbolically; $\gamma = \tfrac12(k+1/k)$ symbolically; and the anchor $t_1 = 2$ , $t_2 = 8$ , $d = 3$ ly, $t = 5$ yr, $\beta = 0.6$ , ping received at her $4$ yr.*

Go deeper · Derivation B · 2 minThe pulse count: differential aging from an integer, with no simultaneity

Derivation BThe pulse count: differential aging from an integer, with no simultaneity

Now the resolution proper. We will compute the two tallies from the $k$ -factor alone and impose the conservation law (3). What comes out is $\tau_{\text{trav}} = T_{\text{home}}/\gamma$ — the standard result — obtained without ever asking what time it is somewhere else. That absence is the point: the twin paradox is usually presented as a puzzle about simultaneity, and this derivation shows you can dissolve it without touching simultaneity at all.

Step 1. Set the protocol. Both twins strobe at proper frequency $f$ — one flash per year of their own life. Let the traveller's total proper time be $\tau$ and Earth's be $T$ ; these are the unknowns, and their ratio is what we want. Why: the protocol is symmetric by construction, so any asymmetry in the counts must come from the physics, not from the setup.

Step 2. Count what the traveller sees. She recedes for the first half of her proper time and approaches for the second — the switch is at her own midpoint, because she is present at the turnaround. So she receives Earth's flashes at rate $f/k$ for $\tau/2$ and at rate $fk$ for $\tau/2$ :

N^{\text{recv}}_{\text{trav}} = f\frac{\tau}{2}\cdot\frac{1}{k} + f\frac{\tau}{2}\cdot k = \frac{f\tau}{2}\left(k + \frac1k\right) = f\tau\gamma ,

using identity (2). Why: this is where $\gamma = \tfrac12(k+1/k)$ pays for itself. The Lorentz factor appears because the traveller spends <strong>equal proper time</strong> in the two Doppler regimes and the arithmetic mean of the two factors is exactly $\gamma$ .

Step 3. Impose the conservation law. Every flash Earth sent was received: $N^{\text{recv}}_{\text{trav}} = N^{\text{sent}}_{\text{home}} = fT$ . Therefore $f\tau\gamma = fT$ , and

\boxed{\;\tau = \frac{T}{\gamma}\;}

Why: that is the whole twin paradox, and it took two lines. No acceleration was analysed, no line of simultaneity was drawn, no turnaround gap was invoked. The traveller ages less because $\gamma$ is the mean of $k$ and $1/k$ and it is bigger than one.

Step 4. Now count what Earth sees, as an independent check. Earth's redshift phase lasts until the turnaround light arrives, at $t_{\text{red}} = L/v + L/c = (L/v)(1+\beta)$ ; the blueshift phase is the remainder, $t_{\text{blue}} = 2L/v - t_{\text{red}} = (L/v)(1-\beta)$ . He receives at $f/k$ then at $fk$ :

N^{\text{recv}}_{\text{home}} = \frac{f}{k}\cdot\frac{L}{v}(1+\beta) + fk\cdot\frac{L}{v}(1-\beta) .

Why: the durations are unequal — this is the light-delay asymmetry, and it is the only asymmetry in the problem.

Step 5. Simplify, and watch $\gamma$ appear again from the other side. Each bracket collapses to the same thing:

\frac{1+\beta}{k} = (1+\beta)\sqrt{\frac{1-\beta}{1+\beta}} = \sqrt{1-\beta^{2}}, \qquad k(1-\beta) = (1-\beta)\sqrt{\frac{1+\beta}{1-\beta}} = \sqrt{1-\beta^{2}},

so $N^{\text{recv}}_{\text{home}} = \dfrac{fL}{v}\cdot 2\sqrt{1-\beta^{2}} = f\cdot\dfrac{2L}{v}\cdot\dfrac{1}{\gamma} = \dfrac{fT}{\gamma} = f\tau = N^{\text{sent}}_{\text{trav}}$ . ✓ Why: the two tallies are consistent — Earth counts exactly the traveller's birthdays, she counts exactly his. The conservation law (3) holds identically, not just at $\beta = 0.6$ .

Step 6. Read the asymmetry off the durations. The traveller's red:blue split is $\tau/2 : \tau/2$ — dead even. Earth's is $(1+\beta):(1-\beta)$ — at $\beta = 0.6$ that is $8{:}2$ . Both twins see redshift at exactly $1/k$ and blueshift at exactly $k$ ; the only thing that differs is for how long. Why: this is the sharpest possible statement of what breaks the symmetry. It is not the shift, it is the schedule — and the schedule is unequal because the news of the turnaround has to travel and the traveller does not. *Verified in verify_aq-sr-twin-paradox-doppler-resolution.py: $N^{\text{recv}}_{\text{trav}} = f\tau\gamma$ and $N^{\text{recv}}_{\text{home}} = f\tau$ symbolically in $\beta$ ; the collapse $(1+\beta)/k = k(1-\beta) = \sqrt{1-\beta^2}$ ; $\tau = T/\gamma$ ; and the anchor tallies $2+8=10$ and $4+4=8$ as exact integers.*

Figure

The Bondi ladder — every flash drawn. β = 0.6, L = 3 ly, k = 2. Earth's ten annual flashes (blue) and the traveller's eight (green) are the only objects in the figure. The turnaround light (gold) leaves the star at t = 5 yr and reaches Earth at t = 8 yr: that 3-year crawl is why Earth's redshift phase lasts 8 years and its blueshift phase only 2. Count the intersections and you have the answer — no simultaneity line is drawn anywhere in this diagram, because none is needed

Limiting cases

Case	Result	Meaning
$\beta \to 0$	$k \to 1$ , both splits $\to 1{:}1$	no shift, no asymmetry, no paradox
$\beta \ll 1$	$k \approx 1 + \beta + \tfrac12\beta^2$	the $\beta$ term is classical Doppler; the $\tfrac12\beta^2$ is time dilation
$\beta \to 1$	$k \to \infty$ , $t_{\text{blue}} \to 0$	Earth sees almost her whole life compressed into a final blueshifted flash
$\beta = 0.6$	$k = 2$ exactly	$\varphi = \ln 2$ ; the cleanest anchor in the subject
$\beta = 0.8$	$k = 3$ exactly	$\gamma = \tfrac12(3 + \tfrac13) = 5/3$ ✓
product $k \cdot (1/k)$	$= 1$ , always	out and back are exact reciprocals; that reciprocity is why the geometric mean of the seen rates is the emitted rate
mean $\tfrac12(k + 1/k)$	$= \gamma$ , always	the arithmetic mean is the Lorentz factor — the identity the count runs on

Worked examples

1 · The full trip at $\beta = 0.6$ , both ledgers

Traveller goes to a star $L = 3$ ly away at $\beta = 0.6$ and returns. One flash per birthday. Count everything.

Step 1 — identify. $\gamma = (1-0.36)^{-1/2} = 1/0.8 = 1.25$ . $k = \sqrt{1.6/0.4} = 2$ . Check with identity (2): $\tfrac12(2 + 0.5) = 1.25$ ✓.

Step 2 — the totals. Earth: $T = 2L/v = 6/0.6 = 10$ yr, so he emits $10$ flashes. Traveller: $\tau = T/\gamma = 10/1.25 = 8$ yr, so she emits $8$ .

Step 3 — her ledger. She turns at her own midpoint, $4$ yr. Outbound she receives at $1/k = 0.5$ /yr: $4 \times 0.5 = 2$ flashes. Inbound at $k = 2$ /yr: $4 \times 2 = 8$ . Total $2 + 8 = 10$ . ✓ That is exactly the number Earth emitted, so she has counted every one of his birthdays — she knows Earth is $10$ years older, and she knew it before she landed.

Step 4 — his ledger. Turnaround light leaves at Earth-time $5$ yr and takes $3$ yr to arrive: he flips at $8$ yr. Red for $8$ yr at $0.5$ /yr: $4$ flashes. Blue for $2$ yr at $2$ /yr: $4$ flashes. Total $4 + 4 = 8$ . ✓ Exactly her emitted count.

Step 5 — the boundary number. $10$ and $8$ . Their ratio is $10/8 = 1.25 = \gamma$ , and it was obtained by counting integers, with no clock synchronisation, no simultaneity, and no analysis of the turnaround. That is the whole resolution, and the only thing that made it asymmetric was that the light took $3$ years to get home.

2 · Radar the ship, get its speed from one clock

Earth pings the outbound ship at $t_1 = 2$ yr and receives the echo at $t_2 = 8$ yr. What is $\beta$ , where was the ship, and when?

Step 1 — identify. $t_2/t_1 = k^2$ , so $k^2 = 8/2 = 4$ and $k = 2$ .

Step 2 — apply identity (2). $\beta = (k^2 - 1)/(k^2 + 1) = 3/5 = 0.6$ . The ship's speed has been measured with one clock, at one place, using only a stopwatch — no distant clock was ever synchronised.

Step 3 — locate the event. Radar distance $= \tfrac12 c(t_2 - t_1) = \tfrac12(6) = 3$ ly. Radar time $= \tfrac12(t_2 + t_1) = 5$ yr.

Step 4 — the boundary number. The reflection event is $(5\ \text{yr},\ 3\ \text{ly})$ — the turnaround itself, and the echo returns at exactly $t_2 = 8$ yr, which is exactly when Earth's redshift phase ends. The ship received that ping at her proper time $k t_1 = 4$ yr, exactly her own midpoint. The radar construction and the pulse count are the same diagram, read in two directions.

Step 5 — sanity. $\gamma = \tfrac12(k + 1/k) = 1.25$ , and $\tau = T/\gamma = 8$ yr. Everything closes.

Going further

Texts. Bondi, Relativity and Common Sense: A New Approach to Einstein (Doubleday Anchor, 1964) — the source · Rindler, Introduction to Special Relativity, 2nd ed., §3.7 · Taylor & Wheeler, Spacetime Physics, 2nd ed., ch. 4 · Ellis & Williams, Flat and Curved Space-Times, ch. 2 (a modern $k$ -calculus treatment).

Provenance. Signature $(+,-,-,-)$ . Every number recomputed in verify_aq-sr-twin-paradox-doppler-resolution.py, including both ledgers as exact integers. Base defect repaired: the AllQFT base for this slug glosses $k$ as "the ratio of emitted period to received period" while simultaneously asserting $k > 1$ during recession — the two statements contradict each other. $k$ is the received/emitted period ratio, which is $>1$ when receding; the base's own next card ( $k_{\text{approach}} = 1/k$ , "received/emitted") uses the correct convention, so the base is internally inconsistent. The convention is fixed and stated explicitly here.

Scope. The Doppler formula itself is owned by `aq-sr-longitudinal-doppler-effect` and recalled. The trip setup and the proper-time resolution are owned by `aq-sr-twin-paradox`. The simultaneity-gap resolution — the one this article deliberately does not use — is `aq-sr-twin-paradox-asymmetry`. That $k = e^{\varphi}$ is the boost's eigenvalue on a null ray is owned by `aq-sr-lorentz-boosts-hyperbolic-rotations`. The four-vector formulation is `aq-sr-doppler-four-momentum`.

Builds on. The Twin Paradox · Longitudinal Doppler Effect · Asymmetry in the Twin Paradox. Leads to. Relativistic Doppler and the Photon Four-Momentum · Transverse Doppler Effect.

Why does the traveller switch Doppler factor at her own midpoint but Earth does not?

Because she is at the turnaround and he is not. The event that changes the sign of the relative velocity happens at her location, so it changes her observations instantly. It happens three light-years from him, so it changes his observations three years later. That is the whole asymmetry, and it is a statement about signal travel time, not about who is "really" moving.

Doesn't the count depend on when each twin starts and stops strobing?

No, provided both strobe over their entire worldline between the two shared events (departure and reunion) — which is the only sensible protocol, since those are the only two events both twins attend. Every flash emitted in that interval is received in that interval, which is exactly conservation law (3). The tally is then fixed, and it is an integer.

The traveller sees Earth's flashes at 0.5/yr outbound. Isn't Earth's clock running at 1/γ = 0.8 of hers?

Both are true, and they are answers to different questions. $0.5$ /yr is what she sees — light delay included, no corrections applied. $0.8$ is the time-dilation factor she would infer after deciding when each flash was emitted, which requires a simultaneity convention. The $k$ -calculus deliberately never makes that decision, which is why it works with $0.5$ and not $0.8$ . The two numbers are related by $1/k = (1/\gamma)\cdot 1/(1+\beta)$ , the extra factor being the classical recession stretch.

Does the pulse count secretly assume something about the turnaround?

Only that it is fast enough to be treated as a point on the diagram — and even that can be relaxed. Nothing in the count cares about the acceleration; if you give the traveller a finite burn, her red and blue phases simply become $\tau_{\text{out}}$ and $\tau_{\text{in}}$ with a smoothly varying $k$ in between, and the integral $\int f\,k(\tau)\,\mathrm{d}\tau$ still totals Earth's proper time. The conservation law (3) does not care how she turned, only that she did.

Why does γ come out as the ARITHMETIC mean of k and 1/k, and not the geometric mean?

Because the traveller spends equal proper time in the two regimes, and equal time at two different rates means you average the rates arithmetically. The geometric mean $\sqrt{k \cdot 1/k} = 1$ is also meaningful — it says that if you averaged the two seen periods multiplicatively you would recover the emitted period exactly, which is the statement that $k$ and $1/k$ are reciprocals. Both means are true; the count needs the arithmetic one.

Could the traveller be fooled into thinking Earth aged less?

Not by counting. At every moment of her outbound leg she is seeing Earth's flashes arrive slowly, and if she stopped there she might well conclude Earth was ageing slowly — and she would be making a claim about simultaneity, which is where the trouble always starts. But she does not stop; she comes home, and the inbound blueshift more than compensates. The final tally is $10$ , and $10 > 8$ . The count is only unambiguous over the whole closed trip, and over the whole closed trip it is unanswerable.

Doppler-Shift Resolution of the Twin Paradox

Where this lives today

The physical picture

The core equations

Counting flashes

Limiting cases

Worked examples

1 · The full trip at $\beta = 0.6$ , both ledgers

2 · Radar the ship, get its speed from one clock

Going further

Before this

Where this leads

Doppler-Shift Resolution of the Twin Paradox

Where this lives today

The physical picture

The core equations

Counting flashes

Limiting cases

Worked examples

1 · The full trip at β=0.6\beta = 0.6β=0.6, both ledgers

2 · Radar the ship, get its speed from one clock

Going further

Before this

Where this leads

1 · The full trip at $\beta = 0.6$ , both ledgers