Jordan Hale: Alex, I have a question for you before we even get into it — have you ever watched someone refuse to believe something that was, like, obviously true?
Alex Mercer: I mean, professionally? Constantly.
Jordan Hale: Right, okay — so hold that feeling, because it's basically the spine of what we're getting into today. Karl Popper, The Logic of Scientific Discovery, 1934. The man who looked at how science supposedly works and found something that, once you see it, you can't unsee it.
Alex Mercer: The asymmetry.
Jordan Hale: The asymmetry. Which is — okay, so you've got a theory. All swans are white. You go out and observe swans. A thousand swans, all white. Ten thousand. Every single one, white. Have you proven your theory?
Alex Mercer: No. You've corroborated it — that's actually Popper's word for it, corroboration — but you haven't proven it. Because the next swan might be black.
Jordan Hale: And if you see one black swan — one — then the whole theory is gone. Conclusively, logically, done. That's the asymmetry: falsification is decisive, confirmation never is. And Popper made that the foundation of what counts as real science — falsifiability.
Alex Mercer: Which brings in the replication crisis, and this is where it stops being abstract. Psychology, biomedicine — large-scale projects have shown that many published findings simply can't be reproduced. Those are, in Popperian terms, black swans. Counterexamples. The theories should be falling.
Jordan Hale: But they're not, you know? Like, the fields keep going. Scientists keep citing the original studies. And Popper's been sitting there since 1934 saying — no, wait, a failed replication is supposed to matter.
Alex Mercer: So the question is whether his framework is a description of science or a prescription for it. And those are very different things.
Jordan Hale: A law of logic, or — what, an aspiration? Something we nod at and then quietly ignore when it gets inconvenient?
Alex Mercer: Yeah. And I think the honest answer is we're still figuring that out.
Jordan Hale: But wait, the 'aspiration' framing — I don't think that's quite right either, you know? Because Popper wasn't describing what scientists do. He was solving something much older than that. Like, he was answering Hume.
Alex Mercer: David Hume. The problem of induction.
Jordan Hale: Right — and Hume's point is almost brutal in how simple it is. No matter how many particular observations you stack up, you cannot — logically, ever — derive a universal law from them. Like, the sun has risen every morning of recorded history, and that still doesn't prove it rises tomorrow. You can't get from 'every instance I've seen' to 'all instances everywhere.' There's always a gap.
Alex Mercer: And Popper's move was to sidestep that gap entirely. Instead of saying 'how do we prove things inductively,' he said — stop trying to prove. Try to break.
Jordan Hale: That's the click, I think. That's the thing. You're not building certainty up from observations — you're just surviving attempts to destroy the theory.
Alex Mercer: Which is what corroboration actually means. And I think people misread this constantly — corroboration is not a weaker version of proof. It's a categorically different claim. A corroborated theory, in Popper's framework, is one that has passed severe tests. Not one that's been shown to be true. It means: we tried hard to break this, and we couldn't. Yet.
Jordan Hale: Yet. That 'yet' is doing a lot of work.
Alex Mercer: It's basically the whole framework, yeah. And this is where the bold conjectures idea comes in — Popper's methodological ideal wasn't just 'be willing to be wrong.' It was more aggressive than that. Propose the most falsifiable hypothesis you can. Make the biggest, most exposed claim. And then try hardest to break it. Not hunt for data that confirms it — actively design tests that would destroy it if it's wrong.
Jordan Hale: Which is — okay, I mean, that's genuinely hard to do. Like, psychologically hard. Because you've just spent years building this theory and now Popper's saying your job is to torch it yourself.
Alex Mercer: To some extent, yeah. But the logic is airtight — confirmation and proof are not just different degrees of the same thing in his framework. They're not even the same kind of thing. Confirmation is basically noise. Falsification is the only move that actually changes your epistemic state.
Jordan Hale: And that came out of The Logic of Scientific Discovery in 1934, right? Like, Popper laid all of this out when he was still in Vienna, responding to the Vienna Circle — who thought the whole game was verification, showing that your statement matches experience.
Alex Mercer: And Popper basically said verification is impossible for any universal claim — you can never close that inductive gap Hume opened. So it's elegant, almost ruthlessly so. Which is probably why it feels slightly too clean. The mechanism is perfect. But when you try to run it through actual institutions, it gets complicated.
Jordan Hale: But that's exactly where Popper takes it further, right? Like, the elegant machinery doesn't just run — it draws a hard line. The demarcation criterion. A theory is scientific if and only if some conceivable observation could, in principle, refute it. That's the whole gate.
Alex Mercer: The demarcation criterion. Yeah. And he gave two canonical examples of what's on the wrong side of it — psychoanalysis and Marxism.
Jordan Hale: Which is — I mean, that still feels kind of provocative, you know? Like, what's his actual charge against them?
Alex Mercer: It's structural. Both theories are built so that any outcome can be absorbed as confirmation. Patient resists treatment? That proves the repression. Patient complies? That also proves the repression. There's no conceivable observation that the theory can't fold into itself. So it's not falsifiable. By Popper's criterion, that makes it non-scientific.
Jordan Hale: Wait — but non-scientific isn't the same as meaningless, right? Like, that matters.
Alex Mercer: Right, and this is the part Popper is actually careful about — he's explicitly not saying psychoanalysis is worthless or that Marxism says nothing true. The criterion doesn't declare non-falsifiable theories meaningless. It just puts them outside science. That's a much narrower claim than people usually attribute to him.
Jordan Hale: Which is — okay, actually no, that's interesting because the Vienna Circle was doing something harsher. Verificationism, right? They said a statement is meaningful only if experience can verify it. Popper thought that was too strong — verification of any universal claim is logically impossible, so you'd end up throwing out half of physics too.
Alex Mercer: Exactly. Popper's move was narrower: don't ask 'can you verify it,' ask 'could you ever refute it.' Demarcation, not meaningfulness. Different question.
Jordan Hale: But here's where it starts bending — and I think about this a lot — applying the criterion consistently produces these genuinely contested verdicts. Like, picture a graduate student in 2005 pitching a string theory dissertation. The theory generates no predictions testable with any instrument that actually exists. Is the committee supposed to say: not science yet? Not science, full stop? That decision has real consequences for the student's career, the department's funding—
Alex Mercer: And that's the friction point. 'Unfalsifiable' and 'not yet tested' look identical from the outside. Popper's criterion doesn't give you a clean way to tell them apart in real time. It's a judgment call dressed up as a logical rule.
Jordan Hale: Science proceeds by elimination, not accumulation — that's the whole Popperian pitch. But the elimination only works if you can actually agree on what counts as a refutation. And apparently we... often can't? And that's before we even get to the part where the replication crisis hits and Kuhn shows up and the whole social machinery of science turns out to be a lot messier than the logic suggests — which, I mean, that's its own thing.
Alex Mercer: Yeah, I think the criterion is compelling — genuinely — but it doesn't stay solved. The logic is clean; the application is where it buckles.
Jordan Hale: But the buckle is the story, right? Like — the replication crisis isn't just 'the application is messy.' It's Popperian logic running in real time, at scale, and it's mostly not doing what Popper said it would. Psychology's large-scale replication projects — we're talking rates somewhere between 36 and 47 percent of published findings actually holding up. That's not noise. That's the majority of the literature potentially wrong.
Alex Mercer: And those failed replications are — in the most literal Popperian sense — black swans. They should be killing theories.
Jordan Hale: But they're not! The fields didn't collapse. The original studies keep getting cited. So what's happening?
Alex Mercer: This is exactly where Thomas Kuhn comes in. And Kuhn's critique of Popper is basically — you're describing a logic, not a science. Real scientific communities don't respond to anomalies by abandoning their theory. They protect it. They say 'the replication used different methods,' or 'there's a hidden moderating variable,' or just — 'the equipment was off.' They adjust auxiliary assumptions instead of touching the core theory.
Jordan Hale: So the theory survives not because the evidence supports it — but because the community finds somewhere else to point.
Alex Mercer: Right. And Kuhn's paradigm shift concept is the alternative account — science doesn't progress through individual falsifications. It progresses through social revolutions. Anomalies accumulate over years, resistance builds, and then — only then — does the community shift. It's social before it's logical.
Jordan Hale: Which means — wait, that means the crisis vindicating Popper and the crisis indicting institutions could both be true at the same time?
Alex Mercer: I think that's actually the sharpest version of the open question. The failure of replication is exactly what Popper said should matter most. But the institutional response — or non-response — is exactly what Kuhn predicted. Positive results get published, get cited, get funding. The negative ones — the disconfirmations Popper would prize — they don't even reliably make it into journals.
Jordan Hale: No way. The thing Popper says is the most epistemically valuable — a failed test — is the thing the system is structurally designed to suppress.
Alex Mercer: That's the hardest edge of it. Popperian discipline requires you to want to be wrong. But the incentive structure — grants, publications, tenure — punishes being wrong. So we haven't failed to apply Popper by accident. We've built institutions optimized for the opposite of what his method demands.
Jordan Hale: Okay but — and I'm genuinely not sure about this — does that mean Popper was wrong? Or just that we refused to do the thing he said we should do?
Alex Mercer: That's maybe the irreducible tension. Popper is normative — here's how science should work. Kuhn is descriptive — here's how it actually works. And the distance between those two isn't a flaw in either theory. It's just... real.
Jordan Hale: Which kind of raises the question — is falsification a philosophical principle at this point, or is it actually a reform agenda? Like, are we doing philosophy of science or are we writing a policy document for academic publishing?
Alex Mercer: I don't know if those are separable anymore. And I think that's genuinely uncomfortable — it means Popper's logic being airtight doesn't tell you anything about whether science will ever actually follow it.
Jordan Hale: You know what I keep thinking about? The guy at the beginning — the one who refuses to believe something obviously true. That was the setup. But now I'm not sure that's even the right frame anymore. Like, maybe it's not refusal. Maybe it's just — we built a whole system that never asked him to believe it in the first place.
Alex Mercer: That's actually a harder problem. Refusal implies the evidence got through and was rejected. But if the incentive structure means the disconfirming result never gets published — the black swan never even makes it into the room.
Jordan Hale: And Popper's been sitting there since 1934 — like, ninety years — and the answer to 'why do we keep needing to be reminded of this' might just be that the logic is airtight and the institutions are not. Those are two different things that we keep treating as one problem.
Alex Mercer: I don't know if that resolves it, though. Because the logic being airtight doesn't save you. The asymmetry is real — one black swan, and the universal claim is gone, that's just valid deduction — but knowing that hasn't moved the replication rate.
Jordan Hale: No. It hasn't.
Alex Mercer: So I think where I land — and it's genuinely uncomfortable — is that Popper solved the philosophical problem. Completely. The asymmetry between falsification and confirmation is, I mean, it's not contestable. But he didn't solve the human problem. And we keep asking the philosophy to do work it was never designed to do.
Jordan Hale: A law of logic that we have to keep rediscovering. That's — I mean, that's kind of a strange thing to sit with.
Alex Mercer: It really is. Alright — I think that's actually where this one ends for me.
