 There are some people who've absolutely mastered Linux, who can make it do anything they want using some sort of pseudo-science. Science has proven time and time again to be a wildly successful approach to obtaining knowledge. Many interesting and important observations were made before the formalization of the scientific method, but it's hard to argue that the practice of science doesn't have some sort of intimate relationship with truth. I mean, yes, people still do argue that. I mean, computers, which is just really supremely frustrating. However, there's another problem in that some individuals have recognized that success and attempt to hang unjustified claims on the coattails of science, representing their ideas as being somehow scientific and justified when, in fact, they aren't. Astrology, alchemy, numerology, phrenology, feng shui, psychics, a whole slew of weird stuff that's supposedly good for your health. There's no shortage of disciplines which people claim are as good as, or even better than, mainstream science, like physics or medicine, at prediction and accessing truth. I mean, we've got evidence that science is really, really good at finding stuff out, and so we want to believe in things that are scientifically justified. But how are we supposed to separate them from all the pretenders? Differentiating science and pseudo-science is a philosophical and practical concern called the demarcation problem. As with most philosophy, it's not as straightforward as it sounds. We have sort of an intuitive sense of what science and pseudo-science are, but trying to make a hard and fast rule to separate one from the other is actually kind of tricky. Sir Karl Popper, a truly prolific 20th century philosopher with some really impressive earlobes, took a stab at the demarcation problem using a principle that he called falsificationism. For Popper, the main thing that separates science from non-science, or pseudo-science, was an emphasis on strict falsifiability. A scientific claim was one which staked its entire reputation on the fact that you might observe something in the world which would totally disprove it, but you never will. Popper chose Sigmund Freud's theory of psychoanalysis as a sort of punching bag of pseudo-science. He said that one might be led to the mistaking conclusion that it was a fantastic theory because it could never really be proven wrong. After all, any human behavior whatsoever could be explained as some combination of id, ego, and superego, and any clinical observation could be interpreted as an expression of sublimation, or repression, or anal fixation, or really whatever the Freudian wanted. However, according to Popper, that actually makes it unscientific. If there's no possible observation which would totally invalidate the theory, if there's no risky statement that the theory makes, which if proven wrong would mean that it's totally bankrupt, then it's not science. This is sort of backwards to how we usually think about these things. The most successful and well-respected scientific theories are powerful because they explain a bunch of different phenomena, because they make positive claims about what will happen given certain starting conditions. But through the lens of falsificationism, the scientific method isn't really about explaining what's true, it's about proving conclusively what's false. In some sense, that's absolutely right. A scientific hypothesis really only has two possible states, wrong, and maybe I mean it worked this time, so hopefully it keeps working that way in the future, I guess, unless somebody proves otherwise. There are some problems with Popper's criteria that we'll get to in a second, but it is an exceptionally poignant observation, and it does cross a whole lot of stuff off of the list. Like, psychics might claim to be able to predict the future, but they never really pin their entire professional reputations on singular predictions. If they get something right, then it's because they're psychic. If they get something wrong, it's only because reading tarot cards is really hard, guys. Falsificationism does a lot for advancing the enterprise of demarcation. And as a bonus, it turns the problem of induction on its head. After all, if science is only in the enterprise of disproving the theories that can't be true in keeping everything else open as a possibility, then there's no induction going on, only deduction. But as I said, there are a few problems with Popper's position. First, it can be interpreted in a way that excludes a whole lot of stuff that we normally think of as science. I mean, Popper himself originally thought that the theory of natural selection wasn't falsifiable, because it kind of sounds like an untestable tautology, like saying, survivors survive. He later retracted this view after being shown the empirical test that had been done on natural selection, revising his statement to say that it was only very hard to test. But he had sort of demonstrated that dismissing something because it seems unfalsifiable can be kind of subjective. Second, and perhaps more importantly, most scientists don't really think that way. I mean, positive results bias aside, there are an awful lot of scientific papers that make positive claims. Claims that something is causally linked to something else. Either there's an unspoken assumption among all scientists that those are just possibilities that haven't been disproven yet, or there's something else going on here. Despite those significant objections, Popper's insight in falsificationism kicked off a whole bunch of inquiry into the philosophy of science, including the demarcation problem. One such line of inquiry was formulated by Thomas Kuhn, whose book, The Structure of Scientific Revolutions, was written in response to Popper and has been an exceptionally influential work. For Kuhn, Popper's falsificationism was only characteristic of somewhat infrequent scientific revolutions, like general relativity or quantum mechanics, not real normal, everyday science. After all, for most scientists, if you get a result that contradicts the existing paradigm, like say, the law of universal gravitation, you don't publish a paper titled Gravity is Wrong. Instead, you publish something that's more like, hey, I found this curiosity. It kind of looks like it's violating the law of gravitation, maybe? According to Kuhn's theory, scientific revolutions only really happen when there are so many observed problems with the paradigm that something absolutely has to give, a period that he describes as a crisis, followed by a paradigm shift. Incidentally, that's not just management jargon. Kuhn's book is where that term first gained popular use. In order to differentiate normal, run-of-the-mill, non-revolutionary science from pseudoscience, Kuhn proposed a demarcation criteria of puzzle solving. If a result, contrary to the existing paradigm, arises in the course of normal science, there's always some confusion that goes along with that. It becomes a puzzle that the researcher has to account for. Maybe the instruments were miscalibrated, or maybe there's an electrostatic effect that's unaccounted for, or maybe gravity does really weird things at small scales. Whatever the case, in real science, disagreement from prediction is always cause for further investigation. On the flip side, in pseudoscience, there's no real problem if a prediction turns out to be false, no puzzle to be solved or investigated. The psychic just says, oh, I must have misread the tarot cards, and then just keeps on saying wacky stuff. Popper and Kuhn both had huge influences on the philosophy of science. They had different approaches and were strongly critical of each other's work, but oddly enough, their theories had more in common than you might think. Whether you use puzzle solving or falsificationism or any of the other numerous philosophical systems that were developed in Popper's wake as a guide for demarcation, it's reasonably clear what sorts of things are science and what sorts of things aren't if you look closely enough and think hard. Adam Savage has his own criteria. Remember kids, the only difference between screwing around in science is writing things down. Many thanks to the Mythbusters for many years of entertainment and education. You guys will be sorely... ...mythed. Do you like Kuhn's or Popper's criteria for demarcation better? Please leave a comment below and let me know what you think. Thank you very much for watching. Don't forget to blah, blah, subscribe, blah, share. And don't stop thunking.