A theory cannot properly be regarded as empirical unless there is some conceivable observation that, if the theory were false, would refute it.
Principle
"unless" = put the negated version of this idea on the Left side of the arrow. (i.e. "if-not")
If it's not the case that there is some conceivable observation that would refute it, if the theory were wrong, then a theory can't properly be regarded as empirical.
No conceivable observation → can't consider that could refute a theory the theory to be empirical
Goal
If any of these answers are concluding, "Thus, it is empirical", we can instantly throw them out. Principles can only prove what's on the right side of the arrow.
If an answer is concluding "this isn't empirical" then we just want to see the answer establish that there's no conceivable observation that could falsify a certain theory.