Discussions over Lorentz-Ether-Theory and SR Simultaneity, Contraction & Expansion
By Wolfgang G. Gasser
Two recent posts (Jan. 2015):
Suppose we have an object
moving away from us at velocity v and
a time difference DT between the object and us.
Our position object -->
This time difference can be interpreted in two very different ways:
1) It can be considered a simple difference between clock values
valid at the same time.
2) It can be considered a concrete deviation from simultaneity.
In the second case, the time difference DT must be taken into
account when determining the distance (at the same time)
between us and the object. But if the velocity v is not constant
during DT, then we have a problem.
"A simple analysis falsifies SR and LET" – 1999-11-04
Robert B. Winn:
"We have a star that begins to emit light at t=0 in its own frame
of reference. We have an observer at rest, observer A, relative to
the star at distance d from the star. At t=0 in the frames of
reference of the star and observer A, two other observers reach
the position of observer A, observer B traveling toward the star at
a velocity of v and observer C traveling away from the star at a
velocity of v."
I think your simultaneity problem is quite interesting.
------o-----------------------:--------- t = 0
star observer A at rest
< - - - - 10 LY - - - - >
Let us assume that the velocity of observer C (frame F') is
v = sqrt[0.9999]c = 0.99995 c --> gamma = 100
relative to the rest frame F. For observer A (frame F) the situation
is very simple. The star begins to emit light at t=0 which will be seen
10 years later.
The Lorentz transform with v = 0.99995 c gives:
 Dx' = 100 (Dx - 0.99995 Dt c)
 Dt' = 100 (Dt - 0.99995 Dx /c)
 Dt = 100 (Dt' + 0.99995 Dx' c)
 Dx = 100 (Dx' + 0.99995 Dt'/c)
Because in frame F, simultaneity (i.e. Dt = 0) and a length of 10 light
years (i.e. Dx = 10 LY) are given, we can calculate the corresponding
time difference in frame F' by using :
Dt' = 100 (Dt - 0.99995 Dx ) = - 999.95 years
This means that a clock of observer C shows 999.95 years less than
a clock on the star when the star begins to emit light, if both clocks
are SR-synchronized in frame F'. Therefore for observer C, the star will
only in the far future (i.e. almost 1000 years later) begin to emit light.
For observer B however, which moves in the opposite direction of C,
the star has already long (999.95 years) ago started to emit light.
The impossibility of such predictions becomes even more obvious
if we assume that observers B and C have only accelerated shortly
before t = 0.
Completely absurd are such predictions within (SR-style) Lorentz-
Ether-Theory. What kind of ether properties could be responsible for
the creation of a time difference of almost 1000 years to an object
only 10 light years away, during a short acceleration phase?
Robert B. Winn:
"We have yet to see any attempt to explain why the Lorentz equations
show a moving observer at two different positions at the time a source
of light begins to emit light."
The general explanation of such problems is rather simple. In a loose
way the problem can be attributed to the ambiguity of 'relativistic
In arithmetic there is a symmetry between addition and subtraction
and between multiplication and division.
In Special Relativity however, there are two different kinds of
symmetry. The symmetrical operation of a division by Lorentz-factor
y is on the one hand a multiplication by y, and on the other hand
of reciprocal time dilation and length contraction) it is the
If a space-time-event A is Lorentz-transformed to A' using velocity
v, then A' can be transformed back to A using -v.
A ----> A' ----> A
Fig.0 Lorentz transform of space-time event A
When transforming a distance, it is not enough to simply transform
two space-time-events A and B (having identical time coordinate
values) to A' and B'.
[A, B] ----> [A', B'] ----> [A, B]
1 m no length 1 m
Fig.1 Lorentz transform WITHOUT simultaneity adjustment,
and distances between events
Because of relativity of simultaneity, at least either A' or B' must
be replaced by a new space-time-event (A" or B") in order to
result in a distance in the new frame (requiring simultaneity).
[A, B] ----> [A', B"] ----> [A, B"']
1 m 1 m / y 1 m / y / y
Fig.2 Lorentz transform WITH simultaneity adjustment
An equivalent variant:
[A, B] ----> [A", B'] ----> [A"', B]
1 m 1 m / y 1 m / y / y
Fig.4 Lorentz transform WITH simultaneity adjustment, and
distances between events
"Coordinate transforms VS space-time transforms" – 2000-01-26
EXPANSION is as present in Special Relativity as CONTRACTION.
:: Every moving object contracted by gamma with respect to the
:: rest frame is expanded by the same factor (again wrt the rest
:: frame) when observed simultaneously in the moving frame. (That
:: this requires constant velocities is a serious problem for SR.)
: If you intermix measurements and observations from different
: coordinate systems you can obtain any sort of nonsense you like.
: But in SR, when you only discuss what a given observer can observe,
: there is never any expansion of a moving object (i.e. the moving
: object is never measured to be larger than its proper size).
It is clear that all the
following variants can directly be derived
from the SR equations:
Dx' = Dx * gamma Dt' = Dt / gamma
Dx' = Dx / gamma Dt' = Dt * gamma
To all of them correspond clearly defined physical situations. If a
fast moving dark object of length Dx'= L sends a light signal from each
end at the same moving-frame time, then the distance between
the two points where the signals are emitted is Dx = L * gamma in
the rest frame. This cannot be denied in a reasonable way!
In an analogous way, the time of a moving system goes faster by
gamma wrt an observer at rest in the rest frame. If this were not
true, then also clocks of a rest frame could not run faster wrt the
time of a particle at rest in a moving frame.
: I have no idea what you are trying to say in your last parenthetical
: remark above.
That you have no idea only shows that you do not fully understand
SR length contraction. If not even you understand such a fundamental
principle of SR, then I must conclude that (almost?) nobody on earth
really understands SR (let alone GR).
:: The EXPANSION can also be explained using LET. [...]
: Yes, one can make the same mistake in LET, and obtain the same
: nonsense. LET is mathematically equivalent to SR, and offers the
: same opportunities to make errors.
I did not make a mistake. Try again:
"The EXPANSION can also be explained using LET. Assume a body
moving at v = sqrt(0.99) c wrt the ether. If its rest length is 10 m,
then it is contracted to 1 m. If we send from the center of the body a
signal to both edges, the sum of both signal paths is 100 m. Whereas
the path in direction of the ether wind is only around 0.25 m the
signal in the opposite direction crosses an ether distance of 99.75 m
before reaching the edge of the body."
Insofar as LET is equivalent to SR it must be able to explain that wrt
a moving observer the ether shrinks by gamma in the same way as
the moving observer shrinks wrt the ether. Because real contraction
of the ether is impossible, the only remaining possibility is an
expansion of the observer by gamma.
"Why the ether would be observable" – 2000-02-02
But I cannot believe that the basic asymmetry in the ether has
:: no effect at all. Isn't there a problem with the intensity of light?
:: Intensity is inversely proportional to the distance square. If the
:: ether is the medium of light, then light intensity decreases with
:: ether distance, doesn't it?
: Yes, when measured in the ether frame. But intensity is also
: proportional to the solid angle subtended by the detector, and
: _that_ is not invariant. Do the computation and you will find
: that competing effects cancel out exactly.
| - - - - - - - o - - - - - - - |
detector source of detector
The sound intensity registered by the two detectors, unlike the
measured frequency, is not independent of the speed of the medium
(the air) with respect to the system.
A spherical light wave is Lorentz invariant:
x2 + y2 + z2 = (c t)2 --> x'2 + y2 + z2 = (c t')2
But I don't think that all segments corresponding to given solid
angles are Lorentz invariant in the same way. We can for instance
divide the original spherical wave into two parts separated by
the circularly propagating wave
x = 0, y2 + z2 = (c t)2
Thinking about stellar aberration is enough to recognize that this
circular wave cannot be Lorentz-transformed to
x' = 0, y2 + z2 = (c t')2
This means that if the number of photons per solid angle is uniform
in the system at rest then it cannot be uniform in the moving system
and vice versa.
In SR it is reasonable to assume that it is always the frame of the
source where the number of photons per solid angle is uniform.
I doubt however that this is a reasonable assumption in an ether
| - - - - - - - o - - - - - - - | -->
detector spherical detector v
Let us assume that this system moves at v = sqrt(0.99) c wrt the
ether and that the distance between the source and each detector
is compressed from 10 m (rest-length) to 1 m in the ether.
Whereas light to the right detector has to travel over 199.5 m in
the ether, the light path to the opposite detector is only 0.5 m.
1 m / (1 - sqrt(0.99) = 199.5 m
1 m / (1 + sqrt(0.99) = 0.5 m
The sound analogy would result in an intensity difference factor of
(199.5 m / 0.5 m)2. So the intensity registered by the left detector
would be (at least) 159 thousand times stronger than by the right
"Simulation of Special Relativity by LET" – 2000-02-05
only theory which is (in principle) experimentally indistinguishable
from Special Relativity is SR itself. No theory based on absolute
simultaneity such as (so called) Lorentz Ether Theory can be fully
equivalent to SR. The SR time transformation t' = gamma * (t - v/c2 x)
is more than a rather arbitrary convention concerning clock
It is clear that the laws of nature themselves do not depend on the
way we synchronize clocks. They depend however on what is REALLY
SR predicts reciprocal length contraction. The only possibility to
simulate this result in LET consists in using the SR simultaneity
If an observer moves at v = sqrt(0.75) c = 0.866 c, then he is
contracted by gamma = 2 with respect to the ether. If he is looking
in such a way that the line
between his eyes is parallel to the
velocity vector, then the distance between his eyes is contracted
from e.g. 8 cm to 4 cm. If ether simultaneity is relevant to this
observer, then he observes that ether distances in the moving
direction are expanded.
The SR simultaneity concept entails that the observer does not
observe with both eyes at the same (absolute) time. If he is moving
to the right, then a subjective instant consists e.g. of ether time t0
when using the left eye and of ether time
t1 = t0 + 12 cm / c = t0 + 0.4 Nanosecond
when using the right eye. During this small interval the right eye
has moved further 12 cm. So the distance between the eyes at an
observer instant is 16 cm in the ether. Only because the distance
between the eyes is actually expanded by factor 2, ether distances
are contracted by the same factor wrt the observer.
But not even this SR simulation is fully equivalent to the corresponding
original SR prediction.
SR simulation SR
object at |-------| |-------| |---|
rest | | |
| | |
distance | | |
| | |
observer |---| |-------| |---|
Fig. 1 Fig. 2 Fig. 3
In Fig. 1 both object and observer are at rest wrt the rest (ether)
frame. In Fig. 2 and 3 the observer moves wrt the rest frame. In the
SR simulation the observer expands, whereas in SR itself the object
shrinks. The two cases are not equivalent because the angles are
different. Such differences result from the fact that distances
perpendicular to the velocity vector do not change.
Insofar as SR and LET are equivalent, they say nothing about nature,
and insofar as they say something about nature, they are not
It is clear that the laws of nature themselves do not depend on
:: the way we synchronize clocks. They depend however on what
:: is REALLY simultaneous.
: Your first statement is clearly true. Your second contradicts the
: first. The meaning of "simultaneous" depends upon how we set our
: clocks, and Nature cannot care how we do that. Nature has no
: "simultaneous", only humans and their clocks do.
What? Nature has no "simultaneous"??? Do you really assume that
clocks are more fundamental than simultaneity? Don't you know
that all actions-at-a-distance of classical mechanics require
absolute simultaneity. Even SR requires (a modified form of)
The statement "the laws of nature themselves do not depend on
the way we synchronize clocks" contradicts "the laws of nature
depend on what is REALLY simultaneous" only if we assume that
what is "REALLY simultaneous" can be nothing more than the result
from (in principle arbitrary) clock synchronization.
:: If an observer moves at v = sqrt(0.75) c = 0.866 c, then he is
:: contracted by gamma = 2 with respect to the ether. If he is looking
in such a way that the line between his eyes is parallel to the
:: velocity vector, then the distance between his eyes is contracted
:: from e.g. 8 cm to 4 cm. If ether simultaneity is relevant to this
:: observer, then he observes that ether distances in the moving
:: direction are expanded.
: Your dredge up a red herring here -- just use LORENTZ'S equations
: and the question of "simultaneity" never comes up.
I do correctly apply the Lorentz equations. The distance between
the eyes of the moving person is contracted by gamma wrt an
observer at rest.
The perception by the moving observer DEPENDS on the fact, that
he uses both eyes SIMULTANEOUSLY. This simultaneity depends
on NATURE and not on an arbitrary synchronization of two clocks
implanted into the eyes.
The correct application of the Lorentz transformation leads exactly
to the conclusions I have drawn.
: Just note the
: coordinates of every relevant event in the ether frame, transform
: USING LORENTZ'S EQUATIONS to the observer's moving frame,
: and you obtain LET's prediction of what the moving observer will
Yes, but if you do this, then you give up the simultaneity concept
advocated by both Lorentz and Poincaré. Don't you see that to
the mathematical coordinates and their transformations must
correspond concrete entities in nature?
: Haven't you noticed: in every article you write you NEVER use
: Lorentz's equations, you always try to apply "length contraction"
: or "time dilation" in one way or another.
Nonsense. It is you who claim that a whole class of theories
(including e.g. Franco Selleri's "inertial transformation" based
on a time transformation entailing absolute simultaneity) are
experimentally indistinguishable from SR and LET.
So it is YOUR claim that "length contraction" and "time dilation"
are the only empirically relevant concepts of SR.
"Fatal error in Lorentz Ether Theory" – 2007-01-22
Adherents of LET often make statements such as "Lorentz ether
theory does not know any paradoxes of relative motion" . Even
some of those who adhere to Special Relativity use LET in order to
defend SR from paradoxes. They argue that LET, based only on time
dilation and length contraction, makes the same predictions as SRT
the seemingly paradox SR simultaneity concept.
The purported equivalence between SR and LET however stems
an astonishingly huge error, exemplified in this text :
"... two mirrors in parallel motion, with a pulse of light bouncing
between them. In this case the motion of the mirrors actually does
diminish the frequency of bounces, relative to the stationary ether
frame, because the light must travel further between each
reflection. Thus the time intervals 'expand' (i.e., dilate). Given this
time dilation of the local moving coordinates, it's
that there must be a corresponding change in the effective space
coordinate (since spatial lengths are directly related to time
intervals by dx = v dt). In other words, if an observer moves at
speed v relative to the ground, and passes over an object of length
L at rest on the ground, the length of the object as assessed by the
moving observer is affected by his measure of time. Since he is
moving at speed v, the length of the object is v dt, where dt is the
time it takes him to traverse the length of the object – but which
"dt" will he use? Naturally if he bases his length estimate on the
measure of the time interval recorded on a ground clock, he will
have dt = L/v, so he will judge the object to be v (L/v) = L units in
length. However, if he uses his own effective time as indicated on his
own co-moving transverse light clock, he will have dt' = dt (1 - v2)1/2,
so the effective length is v [(L/v) (1 - v2)1/2] = L (1 - v2)1/2. Thus,
effective length contraction (and no transverse expansion) is logically
unavoidable given the effective time dilation."
Generally accepted premises of LET are:
- the ether is rigid
- moving time dt' dilates/expands with respect to ether time dt:
dt = dt' gamma
- moving length dx' in direction of motion
contracts with respect
to the ether: dx = dx' / gamma
the basic reasoning exemplified in the above quoted text
implies instead of contraction of moving objects the exact contrary:
contraction of the rigid ether with respect to moving objects. In
LET (insofar as it is not "a clever restatement" of SR), contraction
results from the motion through the ether , so contraction of the
ether with respect to moving objects is impossible.
 The Twin Paradox in Special Relativity and in Lorentz Ether Theory
 mathpages.com, Reflections on Relativity, 1.5 Corresponding States
 "... assuming that the electron, deformable and compressible, is subject to a
kind of exterior constant pressure whose effect ..." Poincaré, June 1905 (see)
"However, if he uses his own effective time as indicated on his own
:: co-moving transverse light clock, he will have dt' = dt (1 - v2)1/2,
:: so the effective length is v [(L/v) (1 - v2)1/2] = L (1 - v2)1/2. Thus,
:: effective length contraction (and no transverse expansion) is
:: logically unavoidable given the effective time dilation."
: What problem do you have with that? I think that the author of
: mathpages is (as usual) correct on the math.
The author is discussing as possible explanations of the Michelson-
Morley experiment: "transverse expansion" versus "longitudinal
contraction" of "the material comprising Michelson's apparatus".
In order to decide this question, he makes a (consistent) reasoning
(starting from time dilation) leading to the conclusion that the
ether contracts wrt (with respect to) a moving object. Using this
contraction of the ether, he decides the open question in favor of
"longitudinal contraction" of Michelson's apparatus (and against
However, the author does not notice, that contraction of the ether
wrt to the apparatus is the EXACT OPPOSITE of contraction of the
apparatus wrt the ether. The only way to get a contraction of the
ether wrt moving objects is an expansion of the moving objects
(and their rulers) wrt the ether. See . But e.g. Tom Roberts claims
(or at least claimed in January 2000) that "there is never any sort
of 'Lorentz expansion'" .
: The earth's atmosphere appears contracted to a muon
and the muon appears contracted to the earth's atmosphere.
However, insofar as the atmosphere appears contracted to the
the muon itself is expanded wrt the atmosphere, and insofar
the atmosphere appears contracted, the muon is expanded.
This reciprocal contraction/expansion in SR is only possible if we
adopt Einstein's simultaneity concept, where v/c2 x of the time
transformation is as 'real' as v t of the x-coordinate transformation:
x' = gamma (x - v t)
t' = gamma (t - v/c2 x)
The claim that v/c2 x is only a convention has no more justification
than the claim that v t is only a convention.
:: Generally accepted premises of LET are:
:: - the ether is rigid
:: - moving time dt' dilates/expands with respect to ether time dt:
:: dt = dt' gamma
: Yes, as measured in a "stationary" system:
: dt' = dt/gamma (a moving clock appears to slow down; thus a
: really moving clock really slows down)
Correct. Slowing down corresponds to EXPANSION of time intervals.
:: - moving length
dx' in direction of motion contracts with
:: to the ether: dx = dx' / gamma
: No, not in a
: dx' = dx/gamma (a moving rod appears to contract; thus a really
: moving rod really contracts)
Here also you are a victim of the confusion (resp. fatal error) I'm
dealing with. According to SR, dx' = dx/gamma means contraction
of rest-frame distances wrt the moving-frame. An ether distance of
e.g. dx = 1 light-year is contracted to dx' = 1 LY / gamma in the
However, you correctly have deduced your wrong interpretation of
dx' = dx/gamma from premises which are assumed to be valid:
(1) In LET
there is neither expansion of moving objects nor
contraction of the ether.
(2) dt' = dt/gamma
= v = dx/dt
From (2) and (3) you conclude dx' = dx/gamma and according to (1)
contraction of the moving object is the only valid interpretation.
So your point of view is no more justifiable than the one of 'rotchm' 
who came to the opposite conclusion:
"Your quoted example is wrong. It should be the other way."
"The author [of mathpages.com] ... has either screwed up or ..."
The existence of contradicting conclusions in a theoretical system is
a clear indication of inconsistence.
:: However, the basic
reasoning exemplified in the above quoted text
:: implies instead of contraction of moving objects the exact contrary:
:: contraction of the rigid ether with respect to moving objects.
: No, in SRT - no matter what your interpretation - the
: reciprocal. That's mentioned in your ref. 2 as well as in your ref. 3
: ("These transformations, ..., must form a group").
Do you mean by "the
observations are reciprocal" that rest-frame
distances contract in the same way wrt a moving-frame as the
corresponding moving-frame distances wrt to the rest-frame? If yes,
do you suggest that this kind of reciprocity somehow follows from
the group-property of the Lorentz transformation? If yes, how?
dx' = dx/gamma (a moving rod appears to contract; thus a really
::: moving rod really contracts)
:: Here also you are a victim
of the confusion (resp. fatal error) I'm
:: dealing with. According to SR, dx' = dx/gamma means contraction
:: of rest-frame distances wrt the moving frame.
: dx = dx'/gamma
is valid in S' only. It's an often made mistake ...
Should this be a con trick in order to obfuscate your error above,
or are you simply confusing dx with dx'?
Your just introduced version dx = dx'/gamma actually has the meaning
you by mistake first attributed to dx' = dx/gamma: a moving rod
contracts wrt to the rest frame.
This new version dx = dx'/gamma is valid if we assume simultaneity
in the rest frame S, i.e. dt = 0:
dx' = gamma (dx –
v dt) --> dx' = dx gamma
--> dx = dx' / gamma
However, the version we
originally have been dealing with,
dx' = dx/gamma, implies simultaneity in S', i.e. dt' = 0:
dx = gamma
(dx' + v dt') --> dx = dx' gamma
--> dx' = dx / gamma
S-simultaneity leads to contraction of S' wrt S (or expansion of S
wrt S') and S'-simultaneity to contraction of S wrt S' (or expansion
of S' wrt S).
:: An ether distance of e.g.
dx = 1 light-year is contracted to
:: dx' = 1 LY / gamma in the moving-frame.
: "ether" and "is contracted" are incompatible,
Lorentz ether theory is therefore inconsistent.
Let us assume a twin-paradox situation where one twin remains at
rest in the ether whereas the other travels with v = 0.5 c to a point
1 LY away from the twin at rest. So an ether distance dx = 1 LY is
passed in dt = 2 years.
From time dilation follows
that the moving twin does not need 2 years
proper time to reach the destination but only dt' = 2 Yr / gamma. So,
if the ether-distance dx from the twin at rest to the destination is
identical wrt the moving twin (i.e. dx' = dx = 1 LY) we get wrt the
moving twin a proper speed of:
v' = dx' / dt' = dx / (dt/gamma) = v gamma
The only way to get a
moving-twin speed of v' = v is the
that the ether distance dx = 1 LY contracts by gamma wrt the moving
dx' = dx/gamma --> dx'/dt' = (dx/gamma) / (dt/gamma) = v
dilation is time expansion. If we combine time dilation with
length contraction, then velocity as length divided by time obviously
cannot remain invariant. Therefore we must combine time dilation
of a moving object with length expansion of the object (equivalent
contraction of the rest frame).
: I think that you misunderstand what the author of mathpages wrote.
Interestingly the LET-SR
confusion, we are dealing with, shows up as
a different incarnation in the article you've pointed to. There you
"Similarly, the clocks will tick at 3/5 of the rate
in rest, making
one tick 0.6 s of duration. Because this is determined relative to
the stationary ether or absolute space, the effects correspond to
If the clock rate of a moving
clock is 3/5 of the rate of rest-frame
clocks, then one moving-clock cycle does not result in 0.6 sec in the
rest frame, but in 5/3 sec. Imagine simply a transversal light clock
moving at v = 0.8 c wrt the ether.
:: So S-simultaneity leads to
contraction of S' wrt S (or expansion of S
:: wrt S') and S'-simultaneity to contraction of S wrt S' (or expansion
:: of S' wrt S).
: In no inertial system does any object appear to be "expanding".
See also. Your assumptions
- real contraction of moving objects wrt the ether
- no (real) contraction of ether distances wrt moving objects
- no (apparent) expansion of moving objects wrt the ether
are inconsistent with the principle of "apparent relativity" (i.e.
apparent/empirical equality of moving frames with the ether frame).
Therefore, I concluded that you and others at least sometimes
confuse "contraction of ether distances wrt moving objects" with
"contraction of these moving objects wrt to the ether". Nevertheless,
the "fatal error in LET" I'm dealing with disappears if we accept:
apparent expansion of moving objects caused by local Poincaré-
There is however a major empirically relevant difference between
SR and LET:
Assume a small spherical homogenous light source at the center
of a sphere with a diameter of
6 m. If the sphere is at rest in the
ether, the light intensity received from the source by the inner
surface of the sphere is obviously homogenous. Now let us assume
that the sphere moves at 0.8 c (and light propagates isotropically)
wrt the ether. Both the light source and the surrounding sphere are
contracted, resulting in analogous ellipsoids. Nevertheless, the ether
drift of -0.8 c (think about aberration) would strongly increase the
light intensity at the rear end at the expense of the front end.
(Received frequencies are the same.)
See also. At least in this case it is
obvious, that length contraction
and time dilation are not enough to simulate the predictions of
: Of course, you
could say that in SRT one has an "expanded" view of
: the moving distance that is at any time next to one's ruler due to the
: apparent contraction of the moving system.
impression is that our positions have converged at least a little.
And such a convergence of different beliefs should be a goal in any
Let us take a 6 m long (rest length) ruler moving at v = 0.8 c in the
as described in your article.
(See APPENDIX below for my treatment/summary of the example)
6 m' ruler à v = 0.8 c
|-------+-------| 6 m rod at rest
|=============================| v = 0
<- - 3.6 m - -> < - - - - - 6 m - - - - - >
<- - - - - - - - - - 10 m - - - - - - - - - ->
In Lorentz ether theory the ether-length dx = 3.6 m of the contracted
ruler corresponds to physical reality. Nevertheless, a local "proper
length" of dx' = 6 m' is ascribed to the ruler. If we assume moving-
ruler simultaneity dt' = 0, we get from the Lorentz transformation:
dx = gamma (dx' + v dt') à dx = gamma dx' = (5/3) * 6 m' = 10 m
This ether distance of dx = 10 m has a very concrete meaning. Let us
assume that along the whole length of the ruler are small flash lamps
next to each other. If they fire all at the same time (S'-simultaneity)
then the distance between the first and the last flash is 10 m in the
However, the flashes do not appear at the same time in the ether
(S-simultaneity): at first appears the flash from the rear end, and
80/3 Nanoseconds later the last flash from the front end. During
these 26.666 ns the ruler moves 6.4 m, so increasing the distance
from the contracted length of 3.6 m to the expanded of 10 m.
Let us further assume that the flash lamps fire at t' = 0 in red, at
t' = 25 ns' in blue, at t' = 50 ns' again in red, and so on. Because
next firing always occurs, when the rear end passes the ether
position where the flash from the front end fired the last time,
this represents a measurement of ether distances using the ruler
length of 6 m'. In the ether this results in a light source apparently
moving at 0.8 c and changing after each a length unit of 10 m from
red to blue or inversely.
So it should be obvious that only insofar (S'-simultaneity) as the
ruler expands from 6 m' to 10 m wrt the ether, ether distances can
contract wrt the ruler, e.g. from 6 m to 3.6 m'.
So in Lorentz ether theory, three basic length-unit changes can be
1) Contraction of objects moving wrt the ether, caused by concrete
2) Expansion of moving objects, caused by different durations
(wrt ether simultaneity) of the movements of the different
points of the moving objects
3) Apparent contraction of the ether, resulting from the expanded
length units of moving objects
The expanded length of a moving object is physically real also
insofar, as it is the sum of the paths of synchronization light
signals from the center to both ends. For an apparent contraction
of ether distances wrt the WHOLE moving object, it is therefore
required that the FRONT END of the contracted object moves
inertially long enough in the ether, so that its distance from the
rear end can increase from L/gamma to L gamma.
Let us assume a ruler with rest length of 2 light years moving at
v = 0.9999995 c (gamma = 1000). So the distances from the center to
each end are contracted to 0.001 LY wrt the ether. Ether distances
passed by synchronization light signals from the center to the rear
and to the front each are:
to_rear = 0.001 LY / (c+v) c = 0.0005 LY
to_front = 0.001 LY / (c-v) c = 1999.9995 LY
The physical expansion in the ether from the contracted 0.002 LY to
2000 LY is caused by constant MOVEMENT of the front, STARTING
the rear after 0.0005 years receives the
and ENDING 1999.9990 years later when also the front receives the
signal. Such expansions by physical movement are a precondition for
apparent contraction of ether distances wrt to moving objects.
As an accentuation of the thought experiment, let us ask what
happens if the 2 LY long ruler performs a circular movement with
an orbit of 2000 LY. In the ether, the length of the ruler is
contracted to 0.002 LY. As measured by the ruler, the orbit must
contract in the same way as local time is expanded, because
otherwise dx'/dt' does not result in v.
Thus the 2 LY' long moving ruler has to ascribe the same length
to the orbit as to itself: 2000 LY / gamma = 2 LY'. This means
concretely: under moving-ruler simultaneity', the ruler is expanded
over the whole 2000 LY of the orbit, and the front of the ruler is
next to the rear at the same time'.
A ruler with a rest length of 6 m passes at v = 0.8 c a rod of the same
at rest in the ether.
6 m' ruler à v = 0.8 c
|-------+-------| 6 m rod at rest
|=============================| v = 0
<- - 3.6 m - -> < - - - - - 6 m - - - - - >
<- - - - - - - - - - 10 m - - - - - - - - - ->
The length dL' = 6 m' of the ruler is contracted by gamma in the ether:
gamma = 1/sqrt(1-v2/c2) = 1/sqrt(1-0.82) = 5/3 = 1.66
dL = dL' / gamma = 6 m' / (5/3) = 3.6 m
Time intervals dt' of clocks fixed on the ruler (x' = const) are dilated/
expanded wrt to the ether time: dt = dt' gamma
Clocks on both ends of the
ruler are synchronized using a light signal
from the center of the ruler. In the ether system, the light signal to
the right clock (in front) needs
1.8 m / (c-v) = 1.8 m / 0.2 c = 30 ns
and passes an ether distance
of 30 ns * c =
The light signal to the
left clock (in the rear) needs only
1.8 m / (c+v) ) 1.8 m / 1.8 c = 10/3 ns = 3.333 ns
and passes an ether distance
of only 3.333 ns * c =
Thus in the
ether, the rear clock is ahead of the front clock by:
dt = 30 ns - 10/3 ns = 80/3 ns = 26.666 ns
Wrt the dilated/expanded time of the moving clocks, the time offset is
dt' = dt / gamma = 26.666 ns * (3/5) = 16 ns'
Let us define t' = 0 as the moment, the right clock of the ruler
passes the right end of the rod (Fig. 4). The left clock then (ether
indicates t' = 16 ns'.
Fig. 3 |=======================|
Fig. 4 |=======================|
Fig. 5 |=======================|
The distance passed by the left clock of the ruler between Fig 1. and
is 10 m. During a movement of 10 m a clock advances (local
dt' = 10 m / 0.8c * gamma = 10 m = (0.24 m/ns) / (5/3) = 25 ns'
In order to determine the proper speed v' as measured from the
moving ruler, we can use Fig. 4 und Fig. 5. The right end of the rod
passes at t' = 0 the right clock of the ruler, and at t' = 25 ns' the
clock. So wrt the ruler, the right rod end moves dL'
= 6 m' in
dt' = 25 ns', resulting in a proper speed of v' = dL'/dt' = 0.8 c.
In order to determine length contraction of the rod at rest wrt to
the moving ruler, we can take the left ruler end of Fig. 1 and the
right ruler end of Fig. 4, both at the same local time t' = 0. The
distance between these two ends is 10 m in the ether. And because
these 10 m corresponds a proper length of 6 m', the 6 m
contracts to 6 m' * (6 m / 10 m) = 3.6 m' wrt the ruler.
The Lorentz transformation today plays the role the principle of
circular motion played in the old (pre-Keplerian) astronomy
"The Thomas E. Phipps cult" – 2014-12-31
: So how does Phipps and other anti-relativists claim Lorentz
: contradicts Einstein? I don't get it!
Still in June 1905, Poincaré described Lorentz' theory in this way (see):
"Lorentz was also led to assume that the moving electron takes
form of a compressed ellipsoid; ...
... and at the same time [one gets] a
possible explanation of the
electron contraction, in assuming that the electron, deformable and
compressible, is subject to a kind of exterior constant pressure
whose effect [travail] is proportional to the variations in volume. ..."
This is deformation of matter due to absolute motion in the ether.
main innovation (unfortunately causing also lots of paradoxes) of SR is
its simultaneity concept.
If somebody thinks that Special Relativity and Lorentz Ether
identical, then I ask: How can absolute movement in the Ether change
: Suffice it to say that
simultaneity can only be determined through
: If there is an ether, then it affects the forces on all material bodies
: and fields. More generally, the ether will also affect the interactions
: between different material bodies and fields. Synchronization is
: one type of interaction. Therefore, the ether has to affect
: The ether gets only a few side
references in 'The Theory of Electrons.'
: Lorentz doesn't do much with the concept other than call some effect
: 'the ether drag'. He did not come up with any formal proof showing
: that the ether wind affects the electrodynamic field. Nor does he
: specify any properties of the ether. Lorentz says that the concept of
: ether may yet find an important utility.
: My delay in responding was not
due to the speed of light, but instead
: is due to the necessity of rereading 'Theory of Electrons'. I have not
: found in the book any mention of ‘absolute simultaneity’.
Not especially mentioning ether or (absolute) simultaneity means
rather agreement with the general view. And this general view did
not put into question the existence of an ether as a medium for light
propagation, or the concept of absolute simultaneity.
Anyway, the central "formal" error of modern physics is the Lorentz
transformation, and a central "philosophical" error is Einstein's
simultaneity concept. So the question whether Lorentz' or Poincaré's
theories before 1905 implied shrinking of the ether with respect to
moving observers is not that relevant.
How do you concretely explain shrinkage of the ether by "interactions
between different material bodies and fields"? With respect to a
spaceship moving at 0.99995 c, an ether distance of 100 light years
is reduced to 1 LY, as the Lorentz factor is 100.
Let us assume that
technology allows (quite instantaneous)
acceleration to 0.99995 c and that during 100 years, every year a
spaceship starts to a planetary system 100 LY away from Earth.
Because of shrinkage of the ether to 1 LY and time dilation, it takes
only 1 year for the crew to reach the target. I ask you:
1. What is
communication transfer time from a ship to a ship having
started one year earlier and to a ship having started one year
2. Does it
make sense for the spaceships to resynchronize their
clocks during journey, and if yes, what concrete resynchronization
do you suggest?
The less you understand, the more you have to believe
With respect to a spaceship moving at 0.99995 c, an ether
:: distance of 100 light years is reduced to 1 LY, as the Lorentz
:: factor is 100.
Lorentz never showed that the ether is shrinking. He showed
: that composite systems shrink. Einstein just used the fact
: that composite bodies shrink.
: Who cares if the ether shrinks?
We measure the dimensions
: of composite bodies, not ether.
In order to get true relativity (i.e. no privileged inertial
motion(s)) the following must be guaranteed:
Ether distances shrink in the same
way with respect to a moving
observer as moving objects shrink with respect to the ether.
If you accept Special
Relativity then you cannot ignore shrinking of
ether distances with respect to shrunken "composite systems".
As I have shown above the shrinkage of the ether (or rest frame)
results from Einstein's simultaneity concept:
is a fundamental property of nature (i.e. more than a
mere synchronization procedure or convention)
2. Simultaneity is defined by the Lorentz transformation
: Let us assume that technology allows (quite
: acceleration to 0.99995 c and that during 100 years, every year
: a spaceship starts to a planetary system 100 LY away from Earth.
: Because of shrinkage of the ether to 1 LY and time dilation, it takes
: only 1 year for the crew to reach the target.
Let us assume that the first spaceship 0 starts in the year 2200 and the
last spaceship 99 in the year 2299. Because of time dilation, a clock in
the first spaceship will show year 2201 when reaching the target in
2300 (ether time). A clock in the last spaceship 99 will show 2300 at
arrival in 2399.
The distance between two neighboring spaceships (e.g. between
spaceships 50 and 51) is 0.99995 LY. Communication between
neighboring spaceships is possible via electromagnetic signals. Such
signals obviously propagate at c relative to the ether (resp. rest-frame).
Let us assume that
spaceship n sends at t = t0 (ether time) a message
to both neighboring spaceships. Let us call the event when back
spaceship n+1 receives the signal EventBack, and when front spaceship
n-1 receives the message as EventFront. In the ether, EventBack occurs at
tB = t0 + 0.99995 LY / (c + 0.99995c) = t0 + 0.5 years
and event EventFront occurs at:
tF = t0 + 0.99995 LY / (c - 0.99995c) = t0 + 20 000 years
With respect to the ether, EventFront occurs
20 thousand years after
EventBack! Yet with respect to the moving frame, these two events are
simultaneous. Only by declaring events simultaneous, happening
however at very different times, light speed can be made the same
in all directions (i.e. in direction to a front spaceship and to a back
spaceship). Otherwise light speed backward would be roughly 20 000
(1.99995 c / 0.00005 c) times higher than in forward direction.
As the whole journey at v = 0.99995 c only takes 100 years, it is
obviously impossible to send a message to a spaceships in front.
Using moving, time-dilated clocks, we get (roughly):
tB' = t0 + tB / 100
= t0 + 0.005 years
tF' = t0 + tF / 100 = t0 + 200 years
So whereas it takes only 1 year to reach the target
message transfer time to a spaceship in front would need 200 years
(resp. 20 000 years ether time).
Only by resynchronizing all clocks of all spaceships (except one), light
speed invariance could be restored. This could be done by adding
(roughly) 100 years to the clocks of spaceship 1, 200 years to 2, … and
9900 years to spaceship 99.
When reaching the target planetary system, this resynchronization by
up to 9900 years obviously must be cancelled again.
Even within the moving spaceships, only clock-resynchronization can
avoid light speed measurements ranging from 0.00005 c to 1.99995 c.
We humans tend to believe what we want to be true, and we want to be
true what we already believe