I am no more fond of PA than anyone, but I suggest reading the documents before assuming they are in the wrong here. LSI concedes that PAs case is in many cases correct here. They did accept and charge for non-wholesale returns that they shouldn't have. LSI are openly admitting they are at least partially in the wrong.
And I interpreted that as LSI admitting that they returned everything that was returned to them, implying they couldn't tell (or couldn't tell easily, so didn't try) who printed/sold which book. I expect the details will come out at trial, if it ever gets there. I am not holding my breath. You could certainly be right, or it may be a technology point that needs to be addressed.
My main sticking point was the idea that they would expect returns to look crappy, which may happen but mostly doesn't. It's not like readers bought them, took them home, read them, and then returned them. Most probably they sat in a stack for a signing, then sat on a shelf untouched for another period of time, then in their virgin untouched state were repacked and sent back (been there, packed the boxes, should have the T-shirt). Where, evidently, it was cheaper for LSI to pulp them and print new ones than to do whatever they needed to do to send the originals to PA.
On the other hand, perhaps a bunch of PA authors had wildly successful signings where they sold dozens of books, which LSI then reprinted and claimed as returns. If so, where are the happy 'I had a great signing' posts on the PA board?