Inter-Blog Popper Wars: The Impotence of Falsificationism — (Note: If this bores the shit out of you, I’m really, really sorry about that.) The replies to my Popper criticisms, and the slow return of my philosophy of science courses from murky recesses of my brain, are deepening my sense that Popperianism is at bottom a skeptical philosophy of darkness, which, despite the enthusiastic rationalist rhetoric of Popperian advocates, shares more with Rorty-like post-modern pragmatism than pro-reason philosophies of light.
Popper is very much one with the positivists in his fixation on formal logic (which I think, of course, is justified like everything else by a kind of induction from experience). In any case, Popper notes that a universally quantified statement is equivalent to the negation of an existentially quantified statement, e.g., ‘All swans are white’ is equivalent to ‘There does not exist something that is both a swan and not white.’ So the discovery of something that is both a swan and white directly contradicts the theory. (If I were writing a song about Popper, I’d call it “Mad for Modus Tollens”.)
Now, a proposition, so it is said, isn’t scientific unless it is falsifiable. ‘All swans are white’ is scientific because ‘Here is a swan that is not white’ would falsify it. However, in order for a proposition to be decisively falsified, the falsifying proposition must itself be decisively true. That is, if ‘Here is a swan that is not white’ has a probability of less than one, then ‘All swans are white’ is defeated only partially.
However, according to Popper, there is no way to assign any positive probability to any proposition. And falsifying propositions must themselves be scientific, and therefore falsifiable. The probability of a basic observation statement like ‘Here is a swan that is not white’ is no greater than the probability of the theoretical statement, ‘All swans are white’. And a critical inquirer, it would seem, is under just as much an obligation to seek falsification of the observation statement (after all, the alleged non-white swan might be a white swan painted black, an animatronic swan, etc.) as of the original hypothesis, because the observation statement turns out to be just another hypothesis. And one can keep going at this forever, trying to falsify any statement that purports to falsify another. So one wonders how we ever get to falsification.
Well, according to Popper, while propositions cannot acquire any degree of confirmation, they can acquire some degree of “corroboration” by passing experimental tests (by not being falsified). The more and severe the tests, the more corroborated the proposition. Popper insists that this isn’t confirmation. You can’t assign a numerical degree of corroboration. You can just very roughly speak of positive and negative degrees of corroboration. (“This hypothesis is really, really corroborated!”) And of course, well corroborated propositions are, strictly speaking, still no more likely to be true than contradictions, but they have (unaccountably) some positive logical standing. So, if ‘Here is a swan that is not white’ is corroborated, then it can falsify ‘All swans are white’.
But how do you know a proposition is corroborated, or corroborated enough to have falsifying power? Well, there is no way. According to Popper, we (or the relevant scientists) just decide. This is what it comes down to. At some point, we just decide that we’re going to accept a statement as corroborated. But there is nothing to instruct the inquirer whether to reject a theory by deciding that a potentially falsifying lower level hypothesis is corroborated or to defend it by trying to falsify the hypothesis.
This seems far from the clear, deductive logic of science that Popper promises us. Indeed, it smells ripe for relativist appropriation. I can just imagine: “Your ethnic group (social class, sex, whatever) may have decided that hypothesis y is corroborated, and refutes hypothesis x. But our ethnic group (social class, sex, whatever) has decided that hypothesis z is corroborated, and refutes hypothesis y, and thus x is secure.” If there is disagreement about corroboration, I guess we can always resort to…. what? War?
For my part, I have not been made to see what is wrong with being certain in seeing mugs on desks, nor in the problem of a proposition becoming more likely true in light of new evidence. Americans tend to be a little perplexed by the enthusiasm for Popper in the Commonwealth. What is it with you guys?