Special relativity: did Einstein's train passenger know his own seat number? I am studying "Introducing Einstein's Relativity" (D'Inverno, Oxford University Press) and I am trying to understand Fig.2.12, page 23. There it goes:

Observer B sends:


*

*a light ray towards Q at time t(R) which bounces back at t(U)

*a light ray towards P at time t(S) which bounces back at t(V)
Observer B concludes that P and Q are not simultaneous, but that anyhow they are equidistant (because light takes equal time intervals RU and SV to bounce back).
The doubt I have is about the need for active illumination of targets for which we want to measure the distance.
Clearly, our observer B is using a radar, or a flashlight, or some other sort of emitting device. I can see that using a radar is radically different from using a binocular, and that telescopes require parallax considerations in order to calculate distances.
But if active illumination is the only way to produce distance estimations then in Einstein's train carriage experiment the observer on the train (who's only receiving light beams) won't be able to judge his own position inside the carriage. Is this correct?
 A: The entire theory of special relativity is built upon the axiom that light speed is constant to any observer in any reference frame. All the rest is just consequences of this assumption.
So yes, Einstein's passenger will see the train like any other train passenger sees their train.
A: Radar methods are not the only way to establish position and time coordinates. One can use Einstein's rods and clocks. For inertial motion in Special Relativity, they lead to identical coordinate assignments. Some folks (like me) prefer radar methods because they are simpler to analyze and practical. I cannot rely on a really long ruler made of some material extended into space. I think the point is this: provide an operational definition of your measurements---don't presume that some everyday-intuitive notion from Galilean physics carries over.
A: There is nothing about relativity that requires active illumination.  Indeed, there is nothing that requires light at all.  It is in some sense unfortunate, from a conceptual point of view, that there is anything that actually moves at c, i.e., it is unfortunate for the understanding of relativity that light exists.  Sure, it was part of the history of the discovery of relativity, and we use it to great value in experiments, but conceptually, if there was no such thing as light, we could still understand relativity just fine, perhaps even better-- because we would then never fall under the misconception that relativity has something to do with a trick of light.  Relativity is about the geometry of spacetime, and any valid means of measuring times and distances will encounter it, even if light is never used at all.  It is equivalent to having light that always moves at the same speed in any frame, but the postulates of relativity can be stated without the existence of light.  If we do it that way, then the postulates could be used to derive the result that if there existed a massless particle, it would always move at c in any frame, but we could have those postulates and not have any massless particle.  An example would be the twin paradox-- even if there were not light, a spaceship moving at close to c would encounter the twin effect.
A: 
if active illumination is the only way to produce distance estimations 

This requirement (cmp. Einstein 1905) stands in the context of thought-experiments, for the purpose of definition of geometric relations (such as distance ratios) and the evaluation of systematic uncertainties of estimations which might be obtained by other means.

in Einstein's train carriage experiment the observer on the train (who's only receiving light beams) 

It is true that the description of Einstein's train carriage experiment (definition of "simultaneity", 1917) concentrates on participant M observing signal (flash) indications of participants A and B (and not the other way(s) around).
But there is explicitly required the determination to identify M as "mid-point" {between} A and B. Arguably this necessitates (for the purposes of unambiguous definition) mutual observations of A, B, and M, namely:
(1) for each signal indication by A, that
 - A saw that B saw that A had signalled, in coincidence with
 - A having seen that M saw that A saw that M saw that A had signalled,
(2) for each signal indication by B, that
 - B saw that A saw that B had signalled, in coincidence with
 - B having seen that M saw that B saw that M saw that B had signalled, and
(3) for each signal indication by M, that
 - M saw that A saw that M had signalled, in coincidence with
 - M having seen that B saw that B had signalled.
(And satisfying these conditions may still not suffice to identify M uniquely as "mid-point between" A and B; additional conditions involving certain additional participants may to be satisfied.)
So:

[...] Is this correct?

Yes.
