Some remarks on Einstein's 1907 paper on Relativity

[harrylin]

I hope that a short discussion of the history of theoretical development is not too much off-topic here; this is a continuation of a discussion started under another, related topic:
https://www.physicsforums.com/showthread.php?t=574624&page=2

The paper under discussion: Albert Einstein, "Über das Relativitätsprinzip und die aus demselben gezogene Folgerungen," Jahrbuch der Radioaktivitaet und Elektronik 4 (1907)

A photocopy of the original paper (in German) can be found online here:
http://www.soso.ch/wissen/hist/SRT/E-1907.pdf

There are also several English translations available:
- "On the relativity principle and the conclusions drawn from it," in The collected papers of Albert Einstein. Vol. 2 : The Swiss years: writings, 1900–1909 (Princeton University Press, Princeton, NJ, 1989), Anna Beck

- H. M. Schwartz, "Einstein's comprehenive 1907 essay on relativity", AJP 45, No.6 and No.9, 1977 (this is a discussion of the paper with an English translation by the author).

Basing myself on the photocopy of the original German document I found the first translation to be quite reliable in general, but the second translation (of which an apparently illegal copy can be found on internet) strikingly unreliable, despite its noble intentions. As a matter of fact, I searched for the original after I first read that translation and had doubts about its contents.

I'm only repeating what Einstein repeatedly says in all his papers, books, and talks.
For example, look at his [..] 1907 paper [English translation by Schwartz]:

He starts off describing how Newton's laws of physics retained their form under Galilean transformation which he specifically called "the principle of relativity". But then the Lorentz's laws of electrodynamics based on "a stationary immobile ether" put that principle in jeopardy because they did not transform intact. And so the search went out to locate the ether culminating in MMX which affirmed the principle of relativity but it had to be based on a different transformation which Lorentz came up with. The only difference between Lorentz's approach and Einstein's is that Lorentz derived the transformation after an experiment forced him to, whereas Einstein starts with the generalized principle of relativity and derives the transformation from a necessary consequence of applying the second postulate. And he specifically states that it is this second postulate that makes the difference between Lorentz's theory and his own theory of Special Relativity when he says:
' It required only the recognition that the auxiliary quantity introduced by H. A. Lorentz, and called by him "local time", can be defined as simply "time." '
This then leads to the statement that the ether "does not fit in" with his theory.

Poincaré, in his 1904 paper, also recounts under the heading of The Principle of Relativity, how Lorentz "valiantly defended" the principle of relativity by coming up with a new transformation.

Now I'm not saying that anyone but Einstein ever formally presented two postulates as the basis for a theory or even that Einstein was claiming that Lorentz had his own two postulates but the net result was the same as if he had and I think it is helpful to point this out.
[...]
Einstein traces that development in Lorentz's ether theory but he never stops calling it a theory different than his own or pointing out that it is his second postulate which make the difference as a starting point and the lack of an ether as an ending point. I count at least seven times in the first column of page 513 where Einstein refers specifically to Lorentz's 1904 theory and two of those times are in contrast to his own theory of relativity. He never claims that they are merely two interpretations of the same theory. In this and other papers, he always shows a contrast to Lorentz's ether theory as a result of his second postulate.

While I fully agree about Lorentz's adherence to the PoR, I don't agree with some other comments and I suspect that the disagreement is in part due to the misleading translation on which those comments are based. In particular, Schwartz has Einstein introduce the main part of his paper as follows:

In what follows it is endeavored to present an integrated survey of the investigations which have arisen to date from combining the theory of H. A. Lorentz and the theory of relativity.

How do unsuspecting readers understand that sentence? Perhaps such readers would tend to think that Einstein there sets out to combine a theory of Lorentz - perhaps the one of 1904 - with "the theory of relativity" - perhaps his own paper of 1905? However, smart readers might wonder why on Earth Einstein would want to do such a thing.

I will welcome confirmation of such a first interpretation, as that would help to stress how subtle errors can cause huge misunderstandings.
Also alternative interpretations (but please, not yet by those who are already better informed!) will be interesting.
More later.

 

[ghwellsjr]

The issue we were discussing in the other thread was whether Einstein referred to Lorentz's ideas as a separate "theory" from his own. I referred to that paper (since you earlier referenced it) as an example where Einstein used the word "theory" when talking about Lorentz ideas (or whatever you want to call them if they aren't a theory) in the same sentence with his own theory of relativity. You complained that I was quoting an inferior translation. So why don't you quote from the translation that you consider to be quite reliable and provide your own translation so that I can see what you are talking about?

The passages in question are the first two paragraphs on page 513 starting with the word "In" of this unreliable translation.

 

[harrylin]

In what follows it is endeavored to present an integrated survey of the investigations which have arisen to date from combining the theory of H. A. Lorentz and the theory of relativity.

- Einstein 1907 according to Schwartz

How do unsuspecting readers understand that sentence? Perhaps such readers would tend to think that Einstein there sets out to combine a theory of Lorentz - perhaps the one of 1904 - with "the theory of relativity" - perhaps his own paper of 1905? [..]

The issue we were discussing in the other thread was whether Einstein referred to Lorentz's ideas as a separate "theory" from his own. I referred to that paper (since you earlier referenced it) as an example where Einstein used the word "theory" when talking about Lorentz ideas (or whatever you want to call them if they aren't a theory) in the same sentence with his own theory of relativity. [..]

Thanks for the confirmation - you interpret those words just as I did the first time that I saw that passage. And it determines the interpretation of the whole section that follows.

However, Einstein did not mention a "theory of relativity" that was combined with a theory of Lorentz (indeed, that is pure nonsense!). At cause is the translator's misreading of "prinzip" with "theorie". The same introduction is translated correctly in The collected papers of Albert Einstein, as follows:

The following is an attempt to summarize the studies that have resulted to date from the merger of the H. A. Lorentz theory and the principle of relativity.

The first two parts of the paper deal with the kinematical foundations as well as with their application to the fundamental equations of the Maxwell-Lorentz theory, and are based on the studies by H. A. Lorentz (Versl. Kon. Akad. v. Wet., Amsterdam (1904)) and A. Einstein (Ann. d. Phys. 16 (1905)).

- Einstein 1907 according to Beck

That is clear and makes perfect sense, although the double reference to Lorentz is potentially confusing. Thus it may be useful to keep in mind that he had earlier identified "the H. A. Lorentz theory" with Lorentz-1895 (also known as the Lorentz Electrodynamics Theory).

 

[Histspec]

Im folgenden ist nun der Versuch gemacht, die Arbeiten zu einem Ganzen zusammenzufassen, welche bisher aus der Vereinigung von H.A. Lorentzscher Theorie und Relativitätsprinzip hervorgegangen sind.
In den ersten beiden Teilen der Arbeit sind die kinematischen Grundlagen sowie deren Anwendung auf die Grundgleichungen der Maxwell-Lorentzschen Theorie behandelt; dabei hielt ich mich and die Arbeiten von H.A. Lorentz (...1904) and A. Einstein (...1905).

I would more literally translate this as:

In what follows, the attempt is made to summarize into a whole the works hitherto emerged from the unification of H.A. Lorentz's theory and the principle of relativity.
In the first two parts of the work, the kinematic foundations as well as their application upon the fundamental equations of the Maxwell-Lorentz theory are dealt with; on that occasion, I relied on works of H.A. Lorentz (...1904) and A. Einstein (...1905).

So the only difference to Schwartz's translation is the word "principle" instead of "theory". But that's no big problem, since in the first sentence, Einstein evidently was alluding to Lorentz's 1895-theory, which actually violated the principle of relativity and thus also the theory of relativity. (Note that until ca. 1910, Einstein used the expression "principle of relativity" instead of "theory of relativity" in most of his publications, so Schwartz's translation makes perfect sense in historical perspective).
In the next sentence, we have the first (and as far as I know) only occasion when Einstein directly referred to Lorentz's 1904 paper along with his own. Does this prove that he thinks that both papers describe the same theory? That's speculation - from the preceding we know, that both papers were dealing with the principle of relativity, not more... And when we follow the rest of this article, we see that Einstein was elaborating on his own two-postulate approach, while Lorentz's method of introducing one auxiliary hypotheses after the other to hide the aether was not highlighted.

So what did Einstein say about Lorentz's alleged "co-discovery" after 1907? There is an interesting interview in 1919, in which he said:

This led the Dutch professor, Lorenz, and myself to develop the theory of special relativity.

However, Einstein wrote in a letter in 1954 that Lorentz was close, yet he didn't reach special relativity: (Letter to C.J.B. Bremer, July 15 1954):

By careful examination of the experimental facts, Lorentz found out that one has to think of the aether as rigid and acceleration-free (contrary to H. Hertz). Newton's space was "materialized" in this way. Though time didn't appear as a problem at first. Yet it became a problem, because it enters as an independent variable (besides the space coordinates) into Maxwell's equations of "empty space", upon which all electromagnetic processes were founded by Lorentz. Now, all would have been satisfying, when it would have been possible to demonstrate the state of motion of the aether ("absolute rest"). The systematic treatment of this problem by Lorentz led very closely to special relativity, because this problem forced Lorentz to transform spatial coordinates and time collectively. That he didn't make that step to special relativity, simply lied in the circumstance, that it was psychologically impossible for him to dispense with the reality of the aether as a material thing (carrier of the electromagnetic field). Those who witnessed this time will understand it.
German: Durch sorgfältige Behandlung der experimentellen Tatsachen fand Lorentz, dass man den Aether als starr und beschleunigungsfrei aufzufassen habe (im Gegensatz z.B. zu H Hertz). Newtons Raum wurde auf diese Weise 'materialisiert'. Die Zeit aber erschien zunächst nicht als Problem. Sie wurde aber dadurch zum Problem, dass sie neben den Raumkoordinaten als unabhängige Variable in die Maxwell'schen Gleichungen des 'leeren Raumes' einging, auf welche ja von H.A. Lorentz alle elektromagnetischen Vorgänge zurückgeführt wurden. Nun wäre alles befriedigend gewesen, wenn der Bewegungszustand des Äthers ('absolute Ruhe') sich hätte nachweisen lassen. Die systematische Behandlung dieses Problems durch Lorentz führte ganz nahe zur speziellen Relativitätstheorie, da ja das Problem Lorentz dazu zwang, die räumlichen Koordinaten und die Zeit zusammen zu transformieren. Dass er diesen Schritt zur speziellen Relativitätstheorie nicht machte, lag einfach daran, dass es für ihn psychologisch unmöglich war, auf die Realität des Äthers als eines materiellen Dinges (Trägers der elektromagnetischen Feldes) zu verzichten. Wer diese Zeit miterlebt hat, begreift es.)

This reasoning for Lorentz's failure to completely arrive at special relativity is in good agreement with Lorentz's own understanding, as written by him in 1914 (except that Poincaré was mentioned, for sources see my earlier post https://www.physicsforums.com/showpost.php?p=3334553&postcount=41):

The formulas (4) and (7) are not in my memoir of 1904. Because I had not thought of the direct way which led there, and because I had the idea that there is an essential difference between systems x, y, z, t and x',y',z',t'. In one we use - such was my thought - coordinate axes which have a fixed position in the aether and which we can call "true" time; in the other system, on the contrary, we would deal with simple auxiliary quantities whose introduction is only a mathematical artifice. In particular, the variable t' could not be called "time" in the same way as the variable t...Poincaré, on the contrary, obtained a perfect invariance of the equations of electrodynamics, and he formulated the "postulate of relativity".

Regards,

 

[harrylin]

[..] So the only difference to Schwartz's translation is the word "principle" instead of "theory". But that's no big problem, since in the first sentence, Einstein evidently was alluding to Lorentz's 1895-theory, which actually violated the principle of relativity and thus also the theory of relativity.

Note also that apparently the label "the theory of relativity" had not yet emerged (that's another, minor deviation by Schwartz). And as you could have noticed from the earlier discussion, his mistranslation produces the twisted misunderstanding that there Einstein tries to combine a theory of Lorentz with the theory of relativity - while in reality he discusses the combination from which the new theory emerged. If you don't perceive the difference, let me put it in math: a+b=c is not the same as a+c=c.

In the next sentence, we have the first (and as far as I know) only occasion when Einstein directly referred to Lorentz's 1904 paper. Does this prove that he thinks that both papers describe the same theory? That's speculation

Good question, but perhaps a bit imprecise. He identifies his summary of those studies with the new theory.

- from the preceding we know, that both papers were dealing with the principle of relativity, not more...

Ehm no. From the preceding we know that according to Einstein both papers of 1904-1905 were dealing with the unification of the principle of relativity and the Lorentz Electrodynamics theory. And it's easily verified that that is correct.

And when we follow the rest of this article, we see that Einstein was elaborating on his own two-postulate approach, while Lorentz's method of introducing one auxiliary hypotheses after the other to hide the aether was not highlighted.

Not surprisingly, knowing Einstein.

So what did Einstein say about Lorentz's alleged "co-discovery" after 1907? [...]

That's not the topic here and I see no allegation of a "co-discovery" here... this thread is about the paper of 1907 - but thanks anyway.
Note that people change their opinions over time, and that for example Lorentz demonstrated a shocking amnesia concerning Poincare at another occasion - but that's again another topic.

 

[harrylin]

Wrap-up. As a reminder an extract from the 1907 paper:

The following is an attempt to summarize the studies that have resulted to date from the merger of the H. A. Lorentz theory and the principle of relativity.

The first two parts of the paper deal with the kinematical foundations as well as with their application to the fundamental equations of the Maxwell-Lorentz theory, and are based on the studies by H. A. Lorentz (Versl. Kon. Akad. v. Wet., Amsterdam (1904)) and A. Einstein (Ann. d. Phys. 16 (1905)).

- Einstein 1907 according to Beck

Also fine is the following on-the-spot translation by Histspec:

In what follows, the attempt is made to summarize into a whole the works hitherto emerged from the unification of H.A. Lorentz's theory and the principle of relativity.
In the first two parts of the work, the kinematic foundations as well as their application upon the fundamental equations of the Maxwell-Lorentz theory are dealt with; on that occasion, I relied on works of H.A. Lorentz (...1904) and A. Einstein (...1905).

And a little further we read:

We now assume that the clocks can be adjusted in such a way that the propagation velocity of every light ray in vacuum - measured by means of these clocks - becomes everywhere equal to a universal constant c, provided that the coordinate system is not accelerated.
[...]
It is by no means self-evident that the assumption made here, which we will call the "principle of the constancy of the velocity of light", is actually realsized in nature, but - at least for a coordinate system in a certain state of motion - it is made plausible by the confirmation of the Lorentz theory [1895], which is based on the assumption of an ether that is absolutely at rest, through experiment [in particular Fizeau's experiment].

- Einstein 1907 according to Beck, with my abbreviation of footnotes in [].

I think that there Einstein gave a fair description of how the works on relativity by Lorentz and Einstein were viewed at that time, and also how the second postulate relates to Lorentz's older 1895 theory.

There remains a point that I have not yet checked, so I'll do that now here:

Originally Posted by ghwellsjr

I count at least seven times in the first column of page 513 where Einstein refers specifically to Lorentz's 1904 theory and two of those times are in contrast to his own theory of relativity. He never claims that they are merely two interpretations of the same theory. In this and other papers, he always shows a contrast to Lorentz's ether theory as a result of his second postulate.

Check: you refer to the last part of section 1, starting with the reference to MMX. But even according to Schwartz, Einstein doesn't refer at least seven times to Lorentz's 1904 theory.
- I count there six references to Lorentz's electrodynamics theory of 1895, which was unsatisfying.
- I also count one reference to Lorentz's paper of 1904, which he used together with his own paper of 1905 as basis for his following summary of the foundations of the new theory.

Regards,
Harald

 

[ghwellsjr]

There remains a point that I have not yet checked, so I'll do that now here:

Originally Posted by ghwellsjr

I count at least seven times in the first column of page 513 where Einstein refers specifically to Lorentz's 1904 theory and two of those times are in contrast to his own theory of relativity. He never claims that they are merely two interpretations of the same theory. In this and other papers, he always shows a contrast to Lorentz's ether theory as a result of his second postulate.

Check: you refer to the last part of section 1, starting with the reference to MMX. But even according to Schwartz, Einstein doesn't refer at least seven times to Lorentz's 1904 theory.
- I count there six references to Lorentz's electrodynamics theory of 1895, which was unsatisfying.
- I also count one reference to Lorentz's paper of 1904, which he used together with his own paper of 1905 as basis for his following summary of the foundations of the new theory.

Regards,
Harald

Here's the full quote, complete with links:

You are confused (perhaps due to a mistranslation): Lorentz's theory of 1895 to which Einstein refers there was not relativistic. In contrast, Einstein discusses there Lorentz's paper of 1904 and his own paper of 1905 as the same theory. Therefore also the bundle "The Principle of Relativity" starts with Lorentz-1904.

Yes, Einstein traces that development in Lorentz's ether theory but he never stops calling it a theory different than his own or pointing out that it is his second postulate which make the difference as a starting point and the lack of an ether as an ending point. I count at least seven times in the first column of page 513 where Einstein refers specifically to Lorentz's 1904 theory and two of those times are in contrast to his own theory of relativity. He never claims that they are merely two interpretations of the same theory. In this and other papers, he always shows a contrast to Lorentz's ether theory as a result of his second postulate.

And this issue started with my first post on the other thread where I said:

Just in case you are asking about competing theories to Special Relativity (and not General Relativity), the wikipedia article on the One-way speed of light provides two, Lorentz ether theory and Edwards' theory. The only difference between these three theories is the differing "second" postulate that they are based on. The "second" postulate has to do with simultaneity, that is, how they arbitrarily decide to divide the time of the round-trip measurement of the speed of light into its two parts, one for the outbound part of the trip and one for the inbound part of the trip. Special Relativity makes them equal. Lorentz ether theory makes them unequal except in one presumed state of rest. Edwards' theory emphasizes that the division is truly arbitrary and let's you pick and choose anyone you want in any particular situation.

Please note that all three theories share the same "first" postulate, the principle of relativity. This adds to the confusion when you ask about "relativity" without specifying what you mean. I'm assuming that you meant "Special Relativity" which is one of the three relativity theories all based on the principle of relativity.

And you responded with:

If Einstein were still alive, he would probably call that Wikipedia article extremely misleading as it turns a fly (different ways of formulating or interpreting the same theory) into an elephant (different theories)...

You are maintaining that Einstein's theory of Special Relativity as presented in his 1905 paper is one and the same theory as the theory Lorentz presented in his 1904 paper and that Einstein pointed this out in his 1907 paper.

I have been maintaining that they are two separate theories but share the same first postulate of the Principle of Relativity and this is borne out specifically and directly by Einstein in his 1907 paper as well as in all his other papers, books and speeches, although maybe not as directly or specifically.

What makes them different theories is that they have a different second postulate which is the assignment of the one-way speed of light by Einstein to any rest state but to only one rest state by Lorentz.

You have been objecting to the inclusion of the word "ether" when referring to Lorentz's theory when Einstein always talks about it when drawing a contrast between his theory and those that went before his.

Einstein also talked in his 1907 paper about how Lorentz's theory evolved. Since he only refers to these as pre-MMX and post-MMX, I had no way of knowing the specific dates that would be applied to the different manifestations. But my point is that in all of them, Einstein called them theories, Einstein referred to them as incorporating ether, and they were all based on the principle of relativity.

Now maybe the point you are making is that if we start with Lorentz's relativity theory, remove the assumption that light travels at c only in a fixed ether, add in Einstein's second postulate, and we will end up with Special Relativity, then I'm in complete agreement because that is what I have been saying all along.

If that is not what you are saying, them please tell me how I have misunderstood what you are saying and what you would say instead.

 

[harrylin]

Here's the full quote, complete with links:
And this issue started with my first post on the other thread [...] You are maintaining that Einstein's theory of Special Relativity as presented in his 1905 paper [..]

The issue here, which was spun off from the thread on alternative theories in order not to hijack it, is purely about what Einstein's 1907 paper states (sorry if the title doesn't make that clear; but I also mentioned that in the other thread). So, please don't try to mix the topic of that other thread into this thread. The other thread is still open.

The discussion here should be useful to serve as reference for current as well as later discussions - our opinions about other papers or theories are off-topic.

Einstein also talked in his 1907 paper about how Lorentz's theory evolved. Since he only refers to these as pre-MMX and post-MMX, I had no way of knowing the specific dates that would be applied to the different manifestations.

Note: as referenced by Einstein, the label "Lorentz's theory" there refers to Lorentz's electrodynamics theory of 1895. A translation of it together with the original text can found here:
http://en.wikisource.org/wiki/Attempt_of_a_Theory_of_Electrical_and_Optical_Phenomena_in_Moving_Bodies

But my point is that in all of them, Einstein called them theories, Einstein referred to them as incorporating ether, and they were all based on the principle of relativity.

No. Lorentz's 1895 theory was incompatible with the relativity principle - even with the "fix" for MMX. You can verify that it wasn't based on it. Oops that is also slightly off-topic; but it may be helpful for correctly interpreting Einstein's 1907 paper.

Harald

 

[ghwellsjr]

Looking again at the unreliable translation, I see that there is a footnote when Einstein first mentions Lorentz's electrodynamic theory to the 1895 paper. But when Einstein says that it was incompatible with the Principle of Relativity, he was talking about the PoR based on the Galilean transformation, because Lorentz had not yet fully developed his transformation, correct?

 

[harrylin]

Looking again at the unreliable translation, I see that there is a footnote when Einstein first mentions Lorentz's electrodynamic theory to the 1895 paper. But when Einstein says that it was incompatible with the Principle of Relativity, he was talking about the PoR based on the Galilean transformation, because Lorentz had not yet fully developed his transformation, correct?

The relativity principle does not depend on transformations but instead, transformations can be developed in order to match the PoR. But indeed, the Lorentz transformation doesn't follow from that theory.
And funny enough the translation that is unreliable at an important place is the more reliable one at another place, see my next post.

 

[harrylin]

There is a puzzling sentence in the introduction, as it suggests to me a double misunderstanding by Einstein in a single sentence, which is unlikely. Perhaps someone else can offer a more probable explanation.

Funny enough, concerning this sentence my appreciation of translations is just the opposite of before: the translation by Beck is erroneous nonsense (about forces that are not states of a substance but things!) and instead I will cite the better translation by Schwartz

Only the idea of a luminiferous ether as the carrier of electric and magnetic forces does not fit in in with the theory presented here; for electromagnetic fields do not appear here as states of some kind of matter, but rather as independently existing objects, on a par with matter, and sharing with the latter the characteristic of inertia.

- Einstein 1907 (Schwartz)

Or, taking the best of both:

Only the conception of a luminiferous ether as the carrier of electric and magnetic forces does not fit in with the theory presented here; for electromagnetic fields do not appear here as states of some kind of matter, but rather as independently existing things that are similar to ponderable matter and share with it the feature of inertia.

- Einstein 1907 (Schwartz+Beck in the mix )

My comments:

- The conception of a material ether does not fit with the theory. However, I think that the concept "luminiferous ether" did not necessarily imply an ether that is made of some kind of matter, but perhaps I'm mistaken?
Alternatively one could propose that he meant the generic concept of a medium - but that is, as far as I can see, not warranted by the works on which he bases his discussion and he later disagreed with the idea that a medium would be incompatible with the theory. Thus, if that is what he meant, it was a blooper.

- The word "only" appears to suggest that his description of electromagnetic fields is notably different from that of Lorentz-1895 to which he apparently referred. However, that description is not in contradiction with Lorentz's description in that paper:

it may be assumed - and this hypothesis I will use in the following - that ponderable matter is absolutely permeable, namely that at the location of an atom, also the aether exists at the same time, which would be understandable if we were allowed to see the atoms as local modifications of the aether.

Thus, why did Einstein write "Only" instead of "Also"? Was that just a mistake or is there an incompatibility that I overlook?

Harald

 

[Histspec]

There is a puzzling sentence in the introduction, as it suggests to me a double misunderstanding by Einstein in a single sentence, which is unlikely. Perhaps someone else can offer a more probable explanation.

Try the following translation:

However, it was demonstrated surprisingly, that it was only necessary to define the concept of time sufficiently precise, to overcome the difficulty discussed before. Only the idea was necessary, that an auxiliary quantity introduced by H. A. Lorentz, which was denoted by him as "local time", can be defined as "time" per se. If one continues to adhere to the sketched definition of time, then the fundamental equations of Lorentz's theory correspond to the relativity principle, when one replaces the transformation equations above, by such ones corresponding to the new concept of time. The hypothesis of H. A. Lorentz and Fitzgerald then appears to be as a necessary consequence of the theory. Only the idea of a luminiferous aether as the carrier of electric and magnetic forces does not fit into the theory laid out here; namely, electromagnetic fields doesn't appear here as states of some sort of matter, but as independently existing things that are equally valid to ponderable matter and share with it the property of inertia.

So?

Alternatively one could propose that he meant the generic concept of a medium - but that is, as far as I can see, not warranted by the works on which he bases his discussion and he later disagreed with the idea that a medium would be incompatible with the theory. Thus, if that is what he meant, it was a blooper.

Well, http://en.wikisource.org/wiki/The_Development_of_Our_Views_on_the_Composition_and_Essence_of_Radiation Einstein was even more radical:

The foundation of the ether hypothesis is the experimentally based assumption that the ether is at rest. The relativity principle states that all natural laws that hold in a reference frame K^{\prime} moving uniformly relative to the ether are identical with those that hold in K, a reference frame at rest relative to the ether. If that is so, we can just as well imagine the ether is at rest relative to K^{\prime}, not K. It is completely unnatural to distinguish the two reference frames K^{\prime} and K by introducing an ether that is at rest in one. A satisfying theory can only be reached if we dispense with the ether hypothesis. Then the electromagnetic fields that make up light no longer appear as a state of a hypothetical medium, but rather as independent entities that the light source gives off, just as in Newton's emission theory of light. As in that theory, space that is free of matter and radiation is truly empty...
Hence, the hitherto prevailing transformation equations in passing from one frame to another moving relative to it rest on arbitrary assumptions. If these are abandoned, the essence of Lorentz's theory or, more generally, the "invariance of the speed of light" principle can be reconciled with the relativity principle

So in both 1907 and 1909, Einstein thought that the relativity principle demands a medium-less or "empty space".

The word "only" appears to suggest that his description of electromagnetic fields is notably different from that of Lorentz-1895 to which he apparently referred. However, that description is not in contradiction with Lorentz's description in that paper:

it may be assumed - and this hypothesis I will use in the following - that ponderable matter is absolutely permeable, namely that at the location of an atom, also the aether exists at the same time, which would be understandable if we were allowed to see the atoms as local modifications of the aether.

Thus, why did Einstein write "Only" instead of "Also"? Was that just a mistake or is there an incompatibility that I overlook?

The conceptual incompatibility is the state of motion of Lorentz's aether. If you look at his 1920 aether speech, Einstein wrote:
* That the only mechanical property Lorentz left to the aether, was its "state of rest".
* That the aether of general relativity emerges from Lorentz's aether by "relativization", that is, removing also the last mechanical property - its "state of rest".
* But this aether may not be thought of as endowed with the quality characteristic of ponderable media, as consisting of parts which may be tracked through time. The idea of motion may not be applied to it.
Thus the very "idea of motion" of the aether is completely banned in Einstein's theory (both special and general), but remains as a hidden entity in Lorentz's theory - this is what brings people to distinguish between "special relativity" and "Lorentz ether theory".

- The conception of a material ether does not fit with the theory. However, I think that the concept "luminiferous ether" did not necessarily imply an ether that is made of some kind of matter, but perhaps I'm mistaken?

Well, he always denied the generic concept of a "mechanical" medium having a "state of motion" - even in his both aether speeches. So as long as "the idea of motion" can be applied to the aether, so long it is "mechanical".

Regards

 

[harrylin]

[..]
Well, http://en.wikisource.org/wiki/The_Development_of_Our_Views_on_the_Composition_and_Essence_of_Radiation Einstein was even more radical:
[..]
So in both 1907 and 1909, Einstein thought that the relativity principle demands a medium-less or "empty space". [...]

Thanks, that is a pertinent point. It supports my impression that his claim was a philosophical opinion (which he rejected later) that did not directly follow from the papers on which he based his summary.

Well, he always denied the generic concept of a "mechanical" medium having a "state of motion" - even in his both aether speeches. So as long as "the idea of motion" can be applied to the aether, so long it is "mechanical".

Regards

I don't follow your explanation, as "matter" isn't identical to "mechanical" - except if it implies that Einstein had simply forgotten Lorentz's rejection of an ether made of matter. As formulated, it appears that he should have written "Also", not "Only".

And was there another ether speech than the one of 1920?
BTW, since you brought it up (but we are not going to discuss that paper in this thread), he explained what he meant with "applying the idea of motion" by comparing the ether to water of which its state of motion is fundamentally undetectable. Apparently he adhered to the "Machian" school of philosophy according to which we may not contemplate unmeasurable qualities.

Anyway, thanks for the suggestions!