Full Idea
Nothing in the world of nominalistically acceptable things could ground or explain the non-identity of the set {A,{A,B}} with the set {B,{A,B}}.
Gist of Idea
What in the real world could ground the distinction between the sets {A,{A,B}} and {B,{A,B}}?
Source
Peter van Inwagen (Existence,Ontological Commitment and Fictions [2003], p.154)
Book Reference
'The Oxford Handbook of Metaphysics', ed/tr. Loux,M /Zimmerman,D [OUP 2005], p.154
A Reaction
[He cites Goodman for this thought] Van Inwagen is offering this to show that the existence of sets is abstract, whereas Goodman was denying the existence of sets altogether. I'm with Goodman. Nice example.