3 ms·

Incredibly misleading article. I'm reminded of the old joke about "It works in practice, but how does it hold up in theory?"

jfengel 6y ago
Well, that is kinda the point. Popper's job wasn't to tell scientists how to do their job, because they were already doing it. Popper's job was to put the scientists' jobs on a firm theoretical footing, because if scientists want VLTIMATE TRVTH then "it works in practice" really isn't sufficient. Popperian falsificationism doesn't succeed in that, which is pretty much old news. (Start with Lakatos and Kuhn and work forward from there.) It's also not really a problem, because philosophy, like science, works as a series of successive approximations. That kind of hypothesis revision works well enough for the scientists to do their work. The problem that that the article describes is that a lot of people don't recognize that. HN is full of people who claim to subscribe to falsificationism, without being aware of the limitations. (Even weirder, they'll often describe themselves as "logical positivists", even though the actual LPs were specifically against falsificationism, and were criticized by Popper... and they quit when they realized he was right.) Even that wouldn't be a real issue, since (as you note) science does work pretty well despite its flimsy theoretical foundation. But it reaches a real problem when people claim both "it works in practice if not in theory" and "falsification is VLTIMATE TRVTH" at the same time. In particular, it leads to a kind of over-reliance on the foundations of science. It misses the kinds of human judgments that go into knowing when a project fails because it was wrong versus failing because you screwed up the experiment. It leads to over-application of incomplete scientific models, where people will claim "... but it's SCIENCE" when they don't know the limitations of the domain. Science is so important, and scientists so good at it, that it's really tempting to dismiss the theoretical problems. The article does an OK job of pointing out some of the issues involved. It's actually pretty interesting stuff: the epistemology of science didn't end in the 1950s and in a lot of ways it's more lost than it has been at any time since Hume. Most scientists have no problem ignoring it, and are probably more productive when they do. But when they don't ignore it, and get it wrong, it can lead to real-world problems.

jfengel 6y ago

Well, that is kinda the point. Popper's job wasn't to tell scientists how to do their job, because they were already doing it. Popper's job was to put the scientists' jobs on a firm theoretical footing, because if scientists want VLTIMATE TRVTH then "it works in practice" really isn't sufficient. Popperian falsificationism doesn't succeed in that, which is pretty much old news. (Start with Lakatos and Kuhn and work forward from there.) It's also not really a problem, because philosophy, like science, works as a series of successive approximations. That kind of hypothesis revision works well enough for the scientists to do their work. The problem that that the article describes is that a lot of people don't recognize that. HN is full of people who claim to subscribe to falsificationism, without being aware of the limitations. (Even weirder, they'll often describe themselves as "logical positivists", even though the actual LPs were specifically against falsificationism, and were criticized by Popper... and they quit when they realized he was right.) Even that wouldn't be a real issue, since (as you note) science does work pretty well despite its flimsy theoretical foundation. But it reaches a real problem when people claim both "it works in practice if not in theory" and "falsification is VLTIMATE TRVTH" at the same time. In particular, it leads to a kind of over-reliance on the foundations of science. It misses the kinds of human judgments that go into knowing when a project fails because it was wrong versus failing because you screwed up the experiment. It leads to over-application of incomplete scientific models, where people will claim "... but it's SCIENCE" when they don't know the limitations of the domain. Science is so important, and scientists so good at it, that it's really tempting to dismiss the theoretical problems. The article does an OK job of pointing out some of the issues involved. It's actually pretty interesting stuff: the epistemology of science didn't end in the 1950s and in a lot of ways it's more lost than it has been at any time since Hume. Most scientists have no problem ignoring it, and are probably more productive when they do. But when they don't ignore it, and get it wrong, it can lead to real-world problems.