On 10/7/20 9:34 AM, Gary Harnagel wrote:
> A true causality violation occurs
> when a message is sent back into the past and is returned to the sender before he
> sent it, and that can only be done in the two frames directly involved with making
> that (purportedly) happen.
Here I explicitly prove that for any tachyonic signal with speed greater
than c there are always simple scenarios in which a "true causality
violation" occurs.
I use no Minkowski diagrams, as there are too many
quibbles about them. This is an algebraic proof based
on Lorentz transforms between inertial frames. I do
not know why none of the major participants in this
thread have not done this, as it is definitive.
This is, of course, a gedanken.
Consider inertial frames S and S' with coordinates (x,y,z,t) and
(x',y',z',t'), having parallel x/x' axes with S' moving in the +x
direction with speed v relative to S; ignore y,z,y',z'. Arrange the
coordinates so x=0,t=0 coincides with x'=0,t'=0. Let g=1/sqrt(1-v^2) and
L be a positive distance. Let w be the speed of the tachyonic signal
relative to the inertial frame in which the transmitter and receiver are
at rest. Use units with c=1, so 0<v<1 and w>1.
At x=0,t=0 a transmitter at rest in S sends a tachyonic signal in the +x
direction, which is received at x=L,t=L/w by a receiver at rest in S.
We have pre-arranged a transmitter at rest in S' to coincide with
x=L,t=L/w, so the receiver in S can instantly send a normal signal to
this transmitter in S' over zero distance. The transmitter in S'
immediately sends a second tachyonic signal in the -x' direction. That
signal was clearly sent at:
x' = g(x - v t) = g L - g v L/w
t' = g(t - v x) = g L/w - g v L
In passing note the following:
* x' > 0 (as expected)
* if 1/w > v: t'>0 and there can be no causality violation.
* if 1/w < v: t'<0 and the analysis continues; this
condition always applies in the limit v -> 1.
* for the example used in this thread (w=10,v=0.5), t' is
clearly negative.
* For the case 1/w < v: if one claims that no signal can
ever be sent in the negative time direction relative to
any inertial frame, that claim is already violated, as
in S' the signal was sent at t'=0 and gets to the S'
transmitter at t'<0.
We have also pre-arranged a receiver at rest in S' to coincide with x=0
when it receives that second tachyonic signal. It will immediately send
a normal signal over zero distance to the transmitter in S, completing a
causal loop -- the question is: does it arrive at t>0 (causality OK), or
t<0 (causality violation)?
Use (X',T') for the coordinates of the second tachyonic signal in S',
and compute (X',T') such that the receiver is located at x=0.
The second tachyonic signal propagates in S' as a function of T':
X' = (g L - g v L/w) - w (T' - (g L/w - g v L))
[This is just a linear trajectory in the -x' direction
starting at the S' coordinates given above.]
The condition for the receiver to be located at x=0 is:
0 = x = g(X' + v T')
Solving these two equations in two unknowns (X',T') gives:
T' = (2 g L - g v L/w - g w v L) / (w - v)
X' = -v T'
Now we can calculate the arrival time T of the signal back at the
transmitter in S (at x=0):
T = g(T' + v X') = g T' (1 - v^2) = T'/g
= [L/(w-v)] (2 - v/w - w v)
As w>v and L>0, the factor in brackets is positive so the sign of T is
sign(T) = sign(2 - v/w - w v)
This means that for any value of w greater than 1 there is a range of v,
2/(1/w+w)<v<1, for which T<0 and causality is violated.
For the example in this thread with w=10: 0.198<v<1,
which includes the example's v=0.5.
Thus, for any tachyonic signal with speed greater than c relative to an
inertial frame, there are simple scenarios in which a "true causality
violation" occurs. It is also true that for such a signal there are
inertial frames in which it propagates in the negative time direction.
Both of those imply that such tachyonic signals are incompatible with SR
and causality.
Tom Roberts