Exercise 3: Why Falsifiability Tests Structure, Not Truth — Possible Solution ================================================================================ Popper's criterion asks a question about a theory's own logical structure: could a real observation, in principle, contradict it? This is a genuinely different question from "is the theory actually true?" - a theory can be perfectly falsifiable and still turn out to be false, and a theory can be unfalsifiable while happening to describe something real that just isn't stated in a testable way. General relativity's real 1919 test demonstrates this distinction directly. Before the eclipse expedition, nobody could yet know whether the theory was true - that's precisely what the test was for. What made the theory scientific, in Popper's own sense, was that it specified a real, exact prediction that the actual observation could have contradicted. The theory's real scientific status didn't depend on the outcome being confirmation; it depended on a genuine risk of disconfirmation existing at all, before anyone knew which way the result would go. If Popper's criterion instead asked "is this theory true," it would need to already know the answer before evaluating anything - making it useless as a criterion applied in advance. By asking "could this be proven false," the criterion can be applied to a genuinely new theory before its real truth is even known, based purely on how the theory is structured. ANSWER: Falsifiability asks whether a real, exact test could show a theory is wrong - a structural property that can be evaluated even before the theory's own actual truth is known. General relativity's 1919 test shows this clearly: the theory was scientific because it specified a real, risky, checkable prediction in advance, regardless of whether that specific test happened to confirm or refute it. Asking whether a theory is simply "true" instead would require already knowing the answer, making it useless as an advance criterion. WHY THIS WORKS AS AN ANSWER ------------------------------ This explains the real logical distinction between falsifiability and truth, using the chapter's own 1919 case to show why the criterion can be applied before a theory's own truth value is established.