"Richard Clark"
Does it say I am the walrus?

No, although another infamous party to this thread bears a striking
resemblance (including precious 'ivory towers' [tusks], thick blubber and a
tough hide)...

http://images.google.com/images?q=Odobenus+rosmarus


To clarify things, lets go out 27.3 wavelengths from antenna A and mark an
X. When antenna A is in receiving mode, the wavefront at X is essentially
flat because antenna B is in the next county. When antenna A is in transmit
mode, the wavefront at X is noticeably curved with a radius of something
around 27.3 lambda. In other words, things are different even if the
pathloss is the same in either direction.

I believe that it would be possible to design an antenna system (an array, a
reflector system, or something with an RF lens) that took advantage of the
difference between 'flat' versus 'curved' to produce an antennas system that
had different pathloss in different directions (where antenna B was
something simpler).

No sense arguing about it. Someone has to produce an example.

The ball is NOT in your court.