Hmm. The argument's close to being circular, as (6) may restate (5). The question would be whether we can know that nothing satisfies the idea since maybe the idea applies to an unobservable entity. You might be assuming empiricism, which would discount unobservables in science. How is it "clear" the idea is empty? You might say, as I do, that the idea is incoherent.