JesseM
Science Advisor
- 8,519
- 17
cos said:Some people may be able to "find inertial frames where the Earth twin's clock ticks slower than the traveling twin's clock during the trip away from the Earth" however this does not comply with Einstein's chapter 4 depiction and it is that depiction to which my posting specifically applies!
It makes no difference if clock A in Einstein's chapter 4 depiction travels a distance to clock B as Einstein shows or if clock A is initially at rest alongside clock B and moves away from B for the same length of time (t) at the same velocity (v).
In accordance with his equation (.5tv2/c2) clock A will lag behind B by the same amount. Clock B (the Earth clock) will not, according to Einstein, lag behind (having, as you suggested, ticked slower than) the traveling twin's clock (Einstein's clock A).
You can certainly analyze the scenario Einstein describes in section 4 of the 1905 paper from the perspective of a frame that's different from the one where A and B are initially at rest (the one that Einstein chooses to label as the 'stationary' frame, although from the context it's clear that this is just for reference, and does not suggest the frame is meant to be 'stationary' in any absolute sense). For example, suppose that in their own initial rest frame, the clocks at A and B are 20 light-seconds apart, and initially synchronized. Now suppose that clock A is instantly accelerated to 0.8c relative to clock B, so that it takes 20/0.8 = 25 seconds to reach the position of B in B's frame. While it moves at 0.8c, in this frame its rate of ticking is slowed down by a factor of sqrt(1 - 0.8^2) = 0.6, so that in those 25 seconds it only advances forward by 0.6*25=15 seconds, meaning it will be 10 seconds behind clock B when it reaches the position of clock B. This is not quite the same as what's predicted by Einstein's formula of (1/2)*t*v^2/c^2, but that's because he earlier approximated (1 - sqrt(1 - v^2/c^2)) as (1/2)*v^2/c^2, "neglecting magnitudes of fourth and higher order". The non-approximate formula would be t*(1 - sqrt(1 - v^2/c^2)).
Now, the point is that there is no obligation to analyze this situation from the perspective of the frame where A and B are initially at rest. You could analyze this same situation described by Einstein from the perspective of a situation where A and B are initially in motion at speed v (which is 0.8c in my example), and then when A accelerates it comes to rest and B continues to move towards it at v. In terms of my example, if A and B were initially 20 light seconds apart in their rest frame before A accelerated, then in a frame where they were initially moving at 0.8c, the distance between them would be shrunk to 20*0.6 = 12 light-seconds due to Lorentz contraction. Also, is A and B were initially synchronized by the Einstein synchronization convention (which Einstein describes in section 2 of the paper) in their own rest frame, then in the frame where they are moving at 0.8c they will not be synchronized, thanks to the relativity of simultaneity--in general if two clocks are a distance L apart in their own rest frame and synchronized in that frame, then in a frame where they are moving at speed v along the axis between them, they will be out-of-sync by a factor of vL/c^2, with the time on the trailing clock being ahead of the time on the leading clock by this amount. So, in the frame where A and B are initially moving at 0.8c, they will be out-of-sync by (0.8c)*(20 light-seconds)/c^2 = 16 seconds. Since we are picking a frame where B is moving in the direction of A, B is the trailing clock here, so its time is the one that's ahead by 16 seconds. So, if A suddenly decelerates and comes to rest in this frame when it reads 0 seconds, B will already read 16 seconds at the "same moment" in this frame. From then on B will be moving towards A at 0.8c, and hence slowed down by a factor of 0.6 in this frame while A now ticks at the normal rate in this frame since it's at rest. Since the initial distance between them is 12 light-seconds in this frame, it will take 12/0.8c = 15 seconds for B to catch up with A. During this time A will advance forward by 15 seconds but B will only advance forward by 15*0.6 = 9 seconds. Since A started out reading 0 seconds at the moment it came to rest, and B started out reading 16 seconds "at the same moment" in this frame, then when B catches up with A, A will read 0 + 15 = 15 seconds, while B will read 16 + 9 = 25 seconds. So, in this frame we get the exact same prediction that A is behind B by 10 seconds when they meet, in spite of the fact that in this frame A was ticking faster than B after A accelerated, not slower. There is nothing about Einstein's thought-experiment that requires us to analyze it from the perspective of a particular inertial frame, we'll get the same final answer to what the clocks read when they meet regardless of what frame we use.
cos said:So according to your previous comment "you can find inertial frames where the Earth twin's clock ticks slower than the traveling twin's clock during the trip away from the Earth" the Earth clock will lag behind the traveler's clock presumably by .5tv2/c2 "then the traveling twin's clock ticks slower than the Earth twin's on the return journey after the turnaround" also presumably by .5tv2/c2 aren't the Earth clock and the traveler's clock going to read the same time?
Again, the 0.5tv^2/c^2 formula is just an approximation. And in this situation as in the previous one, you'll get the same answer to what the clocks read when they meet regardless of what frame you choose. You also have to take into account that if you pick a frame where the Earth is in motion, the time for the traveler to get a certain distance away from the Earth is not the same as the time for the traveler to return back to the Earth from that distance, as it would be in the Earth's rest frame.
For instance, suppose that in the Earth's rest frame, the ship moves away at 0.8c until it reaches a star 20 light-years from Earth (in the Earth's frame), then turns around and returns to Earth at 0.8c. In the Earth's frame, both the inbound leg and the outbound leg will last for 20/0.8 = 25 years, and the traveler's clock will be slowed down by a factor of 0.6 on each leg, so the traveler will only age by 25*0.6 = 15 years on the outbound leg, and will age another 25*0.6 = 15 years on the inbound leg, so when they reunite the Earth twin is 25 + 25 = 50 years older while the traveling twin is only 15 + 15 = 30 years older.
Now let's look at this from the perspective of a frame where the Earth is moving at 0.8c, and the traveler is at rest during the outbound leg, letting the distant star come to him, then when the star reaches him he accelerates in the direction of the Earth, moving towards the Earth at (0.8c + 0.8c)/(1 + 0.8*0.8) = 1.6c/1.64 = 0.975609756c (using the formula for relativistic velocity addition). If the star was 20 light years from Earth in the Earth rest frame, in this frame the distance between Earth and the star is only 20*0.6 = 12 light years due to Lorentz contraction. So when the traveler first comes to rest near the Earth, and the Earth moves away at 0.8c while the star moves towards him at 0.8c, it will take a time of 12/0.8 = 15 years for the star to catch up with him in this frame. At the moment the star catches up to him, the Earth is now 12 light years away, and then the traveler accelerates to 0.975609756c in the direction of the Earth, while the Earth continues to move away at 0.8c. So, the distance between them is only shrinking at a rate of 0.975609756c - 0.8c = 0.175609756c, which means it will take a time of 12/0.175609756 = 68.33333 years for the traveler to catch up with the Earth again, in this frame. During this return phase, the traveler's aging is slowed down by a factor of sqrt(1 - 0.975609756^2) = 0.2195122, so he'll only age by 68.33333*0.2195122 = 15 years during this phase. Meanwhile, during both the first phase and the second phase the Earth was always moving at 0.8c in this frame, and so the Earth's clock was always slowed down by a factor of 0.6, so during the first phase the Earth-twin aged 15*0.6 = 9 years, and during the second phase the Earth-twin aged 68.33333*0.6 = 41 years, and so the Earth twin aged a total of 9 + 41 = 50 years from the time the traveling twin departed to the time the traveling twin caught up with Earth again. So you see, this frame ends up predicting exactly the same thing about their ages when they reunite as was predicted in the Earth rest frame--this frame predicts that when they reunite, the Earth twin has aged 50 years while the traveling twin has only aged 15 + 15 = 30 years. That's despite the fact that in this frame, during the first phase of the the trip the Earth-twin aged less than the traveling twin, with the Earth-twin aging only 9 years from the time the traveling twin departed to the time the traveling twin turned around (again, relative to this frame's definition of simultaneity), and the traveling twin aging 15 years between departure and turnaround.
Nope, Einstein would disagree--you're just failing to understand the relativity of simultaneity (which Einstein briefly discusses at the end of section 2 of the 1905 paper, and which he also explains in more detail in sections VIII and IX of this book), which means that if A and B were synchronized in their own rest frame, they are out-of-sync in other frames. So, in a frame where they're out-of-sync, the time on B may be significantly ahead of the time on A at the moment that A changes speed, meaning that even though B is ticking more slowly as it approaches A in such a frame, it will still be true that B is ahead when they meet.cos said:According to Einstein's chapter 4 depiction there is an "objective truth about whose clock is ticking slower at any given moment". According to Einstein it is clock A that is ticking slower than B at any given moment.
Last edited: