Max Jammer has written a "comprehensive" survey of Concepts of Simultaneity - From Antiquity to Einstein and Beyond (2006), most of which is in fact devoted to Einstein's definition of simultaneity and subsequent debates relating to it. On reading this book, it is striking that apparently those debates have almost exclusively centred on the question of whether or not Einstein's clock adjustment procedure is the only possible way to synchronize clocks rather than whether or not it is a synchronization procedure in the first place. Attentive readers will have noticed that in my view the latter question is much more relevant for an understanding of Special Relativity and the principle of the constancy of c.
In particular, if Jammer is to be believed, the entire global community of physicists and philosophers seems to have completely ignored the requirement that the conditions in which an Einstein clock adjustment signal is emitted or propagates must be symmetrical in opposite directions. With one notable exception: one fairly prominent physicist did realize the importance of this requirement and wrote it down a few years after Einstein published his famous 1905 article. His name was... Albert Einstein!
In fact, as Jammer's book shows, Einstein was perfectly aware of the need for symmetrical conditions of signal emission and propagation for his clock adjustment procedure to qualify as a synchronization procedure. In an article on "The Principle of Relativity and its Consequences in Modern Physics" first published in French in 1910, and quoted in Jammer's book on page 124, Einstein writes that the "means of sending signals" in his clock adjustment procedure "must be such that we have no reason to believe that the phenomena of signal transmission in the direction AB differ in any way from the phenomena of signal transmission in the direction BA".
In view of the importance of this quote, I consulted The Collected Papers of Albert Einstein - Volume 3 (1993), in which the original article in French, "Le principe de relativité et ses conséquences dans la physique moderne", published in Archives des sciences physiques et naturelles 29 (1910) pp. 5-28 and 125-144, is reproduced. And I translated the passage myself since the English translation quoted by Jammer inaccurately says "should be such" when the French clearly says "doit être tel" - "must be such" (p. 25).
So, at least according to the French translation of the original manuscript, in Einstein's view this is an essential condition, a "must". Well, I agree... but unfortunately Einstein, having seen the light, fails to carry through this thought to its logical conclusion. Far from realizing that, if the condition is fulfilled in one frame of reference, it cannot possibly be fulfilled for the very same light signals in a second frame of reference that moves relative to the first, Einstein argues that the equivalence of the conditions of light emission and propagation in all directions in any frame is true "by definition" since "the principle of the constancy of the speed of light" ensures that "in empty space, light always propagates at the speed c" (p. 26).
This argument, which is akin to a line of reasoning put forward by Max Born as set out in my previous post, is flawed because a synchronization procedure must be established before a principle such as the constancy of the speed of light in opposite directions even makes sense. Of course, it may be possible to adjust clocks in such a way that, in purely formal terms, the equality of the speed of light in different directions is ensured for every observer, and this is indeed what Einstein did. It remains that such a clock adjustment procedure is not necessarily a synchronization procedure, which logically has to be established before meaningful, rather than purely formal, statements about the constancy of the one-way speed of light in different directions can be made.
After all this, little remains to be said about Allan Janis's online article on the Conventionality of Simultaneity. Apart from a brief discussion of possible causal anomalies in Special Relativity, essentially that article, too, discusses the issue of whether or not Einstein's clock adjustment procedure is the only possible way to synchronize clocks, rather than whether or not that procedure is a synchronization procedure in the first place.
Perhaps the apparent near-absence of debate on such a crucial aspect of the constancy of the speed of light for every observer, and thus of Special Relativity as a whole, should make me stop in my tracks. If all physicists and philosophers agree that Einstein's clock adjustment procedure is a synchronization procedure, who am I to question it? Are physicists not free to synchronize clocks as they see fit, even if that means departing from previous concepts of simultaneity, which they may have come to find unworkable or inconvenient? The answer will have to wait until my next post.
