I think this is what I would do:
1. If I don't have a suitable lens to lest against, buy 2 copies on credit, compare and return one. You might even tell the retailer what you are going to do. This assumes that you/they haven't got a defective 'batch'.
2. Test against a suitable lens you have or can borrow. For instance if you have a lens with a FL of 50-100mm, decide how sharp is should be compared to that lens for you to be satisfied, then test. Maybe it will be a good copy, but still not be what you were expecting?
I totally agree that it's hard to say whether a lens is 'sharp' without direct comparison, although with some experience you should be able to ascertain whether it is sharp enough for you. But again it's too easy to be mislead without a direct comparison IMO.
Bad copies, however, will look obviously unsharp compared with something comparable, even if the FL is not quite the same. Or they will have obvious sharpness variations across the frame (don't pixel-peep the corners at large apertures though - there will likely be some small variations).
I must admit though, for the first time ever I recently bought a second copy of a lens (TS-E 17mm) because I was seeing some effects that I couldn't explain, nor find any info about. Turned out to be exactly the same with the second lens, but hearing all these horror stories does make one a little nervous.
I also had an obvious with my 17-40 when I moved to FF which I never noticed on a crop soft on one side) and after spending $200 and sending it to the shop twice I would say it is 'acceptable', but not great.
But I also now have enough lenses to test anything I'm going to get so see if it meets my expectations.