primitive thisness

my friend Gaurav presented me with an argument from Robert Adams:

Consider two spheres, call them A and B, and possible worlds W,

(1) in W1, there are two distinct spheres A and B that are identical in all their properties except that A has some contingent property p,

(2) in W2, there are A and B where everything is the same as in (1) except that B and not A has some contingent property p,

(3) in W3, only A exists, but without p,

(4) in W4, only B exists, but without p,

(5) in W5, both A and B distinctly exist, but neither having p.

The upshot of the argument is that objects aren't individuated in virtue of their properties, but that individuation is not reducible, or primitive. But if you like Leibniz's law, which basically says you can't have two distinct objects with identical properties, you shouldn't like this argument. i don't know what to say about this argument. It feels like there's some sneaky move being made but i can't quite say what. Any thoughts?