Consider: "The Idea That a Scientific Theory Can Be ‘Falsified’ Is a Myth - It’s time we abandoned the notion", By Mano Singham, 7 Sept 2020, in Scientific American, the world's leading popular science magazine.
Singham argues that falsification in Science is a bad idea. He does not actually give a single concrete example where falsification is a bad idea. Instead Singham argues generalities in the abstract. Almost as if he doesn't have the first clue what science is. Science is an empirical enterprise with the aim of describing the world in terms of the things of the world. Singham's article misrepresents science, what it is and how science works; and it misrepresents what "falsification" means. Critics of science need to first understand the thing they criticise before their criticisms can be taken seriously.
Falsification and validation can not be added or subtracted to/from science, as Singham seems to think. They are science. Science is its method. Or rather, the method of science, and the ideas produced by the method are the same thing to the extent we can't have one without the other. This method was arrived at, by trial and error, over thousands of years. The idea of science began about 2,600 years ago in Ionia (Western Turkey, on the shoreline of the Agean) then inhabited by Greeks; due to two Greek thinkers: Thales and Anaximander. Yet it wasn't until 400 years ago that science took off after it adopted empiricism entirely. For the 2,200 years, prior to the scientific revolution, science was in limbo; going nowhere. After adopting, and improving upon, the scientific method, it made the world what it is today.
But the field known as science studies (comprising the history, philosophy and sociology of science) has shown that falsification cannot work even in principle. This is because an experimental result is not a simple fact obtained directly from nature. Identifying and dating Haldane's bone involves using many other theories from diverse fields, including physics, chemistry and geology. Similarly, a theoretical prediction is never the product of a single theory but also requires using many other theories. When a “theoretical” prediction disagrees with “experimental” data, what this tells us is that that there is a disagreement between two sets of theories, so we cannot say that any particular theory is falsified.
He's wrong. A temperature reading, pressure reading, ..., is a fact. It was observed. It does not matter whether we used instruments such as our eyes, or two instruments: our eyes with a telescope. A fact is a simple statement about the world. In much the same way as a telescope is an instrument, so are your eyes, and all our senses. Collectively, we can all agree that we observe the world in the same way. We both see the same colour when we see red. When we look at the sky at the same time, from the same place, we both see the same stars, in the same place. We don't need 'theories' of the world, scientific or otherwise, to agree that an apple is an apple.
To take Singham seriously we need to prioritise speculation and make believe ahead of empiricism. To do so is to abandon common understanding, empiricism, and common sense. Science gives us an empirical, objective, common understanding of the world. We abandon that at our peril. If we cannot agree that a scientific instrument (telescope), or even our eyes, shows the same thing to consciousness, in the same circumstances, to different people, with diverse belief systems, then we'll never agree on anything. Nor do we need agree on the same 'theories' to reach such a common understanding. Our apple is not theory dependent. It's either there or not.
Singham's argument against falsification also applies to validation. So he's not ditching part of the scientific method. He'd ditching all of it! Yet he can't even see what he's doing. He doesn't know that his argument against falsification is also an argument against all scientific method.
So what advantages do we get by taking Singham seriously?: Any at all? His' is not an argument for a postmodern science. Actual working results using pomo scientific practice by whatever methods or 'anti-methods' postmodernists can agree on would be an arguement for postmodern science. It's at least 50 years since postmodernism and pomo relativism burst on the scene. My request to Singham is simple: show me your alternative science. He has nothing. There are only words published in the "Science Studies" academic niche and fashion of the day in academia, where Singham makes his career gaslighting young people with epistemic relativism.
Unfortunately, some scientists have disparaged the entire field of science studies, claiming that it was undermining public confidence in science by denying that scientific theories were objectively true.
That's not the problem here. Science does not need us to have confidence in itself, no more so than the sun needs us to be confident in itself to rise and shine. The world happens and does its thing no matter what we think about it. Scientific results are true to the extent they accurately represent and predict the world. Scientific results are false, or pseudoscience when they do not represent and predict the world. This is why we have validation and falsification. Both are terms for empirical tests of scientific ideas (hypotheses and theories). Empirical tests are careful observations or experiments. When an idea claims to be about the world but cannot be tested empirically, then it is pseudoscience. Like, for example, String Theory. Validation and Falsification are a sift to filter science from pseudoscience. What does Singham suggest we replace our current sift with? Who knows, but I do know Singham's replacements, will not 'work'.
"That is not only not right; it is not even wrong"
- Wolfgang Pauli
In German: "Das ist nicht nur nicht richtig; es ist nicht einmal falsch!"
The only way to know whether a scientific idea is right or wrong is to empirically test it. If it is impossible to test it is not even wrong. As a scientist aquaintance of mine once told me: "Science just works" Meaning: It's true because it works, or rather: it's true when it works. Whether particular theories work is done by validation and attempted falsification.
Falsification
Karl Popper codified falsification and explained why it was key to science. Popper was a very practical down-to-earth guy. Not a "pie in the sky" philosopher. He was a practical philosopher. Popper didn't invent or discover it, falsification was already in use; since the start of the scientific revolution.
Popper wanted to explain why some scientific ideas (theories) were better than others. His stress on falsification and validation must be understood in that context.
A: In terms of validation both are quite good. In many respects Ptolemy's theory is better than Copernicus! It often fits the data better.
A validation is a test - an empirical (experiment or observation) to compare reality, against what theory predicts. The more successful validations a theory has, and the closer the fit between reality and theory, the more confident we are of the theory correctly explaining things. At that point in time, 412 years ago, Ptolemy's model had loads of good validations.
In contrast a falsification is also a test - an experiment or observation designed to test potential weak points in a theory. We try to find data which specifically refutes some theory prediction. This is based on the idea that a good theory is like one aspect of a simulation of the real world. So if the theory is wrong in any respect, it fails to simulate the world at that point - it's a bad simulation. A bad simulation is a bad theory - it is wrong.
Q: Why is a theory - such as a climate model - bad when it seems to explain the world in totality - but gets some specific details wrong? A: Because the model claims to be a simulation. A simulation is as weak as its weakest point. if some aspect of the simulation is wrrong, the matrix doesn't just have a glitch in it, the matrix is a con!
Imagine, the simulation, such as a climate model, is supposed to work like sequences of events: A → B → C → D. If something is wrong at B, it follows that it must also be wrong at C, and D. Because C happened after B, and uses data from B. Unless, for example, something is ALSO wrong at C to correct the situation! But then we have a simulation: A → B → C → D with two errors: B, and C. The actual error B (bad simulation), and the worse error C (boondoggle to correct B!). Alternatively: with 3 errors: B, C, D (if we don't correct the error at B). My point: the whole simulation must be right for it to predict anything. The final prediction, cannot be right unless every step made to arrive at the prediction is right. Find one fault in it, and the whole is faulty.
Ptolemy's planetary model was chock-a-block full of boondoggles. Copernicus's not so much. Occam's Razor is a good principle. "when deciding between two explanations for something - best go with the simplest"
“It can scarcely be denied that the supreme goal of all theory is to make the irreducible basic elements as simple and as few as possible without having to surrender the adequate representation of a single datum of experience.” Albert Einstein, 1933.
Often paraphrased as “Everything should be made as simple as possible, but no simpler”
A famous falsification happened, in December 1609 and January 1610, when Galileo looked at Jupiter through his telescope and noticed 4 objects revolving around Jupiter. These were 4 of its most visible moons. Galileo reasoned that if Ptolemy's crystal spheres were 'up there', the objects would not be able to move as they did. Objects were not allowed to pass through the crystal spheres, and nothing could move within them. The crystal sphere's were literally holding the sky in place above our heads. Galileo falsified Ptolemy's planetary model by showing that 4 things were moving through the crystal sphere holding Jupiter in place. This paved the way for Kepler and Newton to produce a better model - which lasted another 200+ years until Einstein published his General Relativity in 1915.
By now, an astude reader will have noticed my 'A → B → C → D' schema with an error at B could be Ptolemy's geocentric, crystal spheres model of the celestial bodies. The Ptolemic system only worked because it had kludges in place to correct epi-cycles. Epi-cycles are cycles within cycles seen in the sky which happen when a earth overtakes a planet in orbit around the sun. Hence my reference to Occam's Razor. The Copernican solar-centric model is simpler. Kepler's model was an improvement on Copernicus; it explained all the data at that time & but kept things as simple as possible. Falsification was a key step in arriving at this better understanding of the solar system.
Singham implies that falsification and validation are just ideas. That we can drop how and why we do scientific tests for some other method. No. Falsification and validation are instrinsic to the scientific method, which itself is indistinguishable from science.
Climate modellers are also at war with falsification because several observations falsify their atmospheric model of the greenhouse gas effect. In general, the Left don't like science and facts when it contradicts socialism & woke. For example:
- Gender theory - "transwomen are women", TWAW. They literally do not agree that there are only 2 human sexes: male and female; and that, being biologically determined by DNA, sex is immutable. No one can change sex. Perhaps we can change gender?, but sex is immutable.
- Climate science - "the greenhouse gas effect (man-made global warming) is settled science"
The left seem to slide ever closer to not just accepting epistemic relativism, but mandating it! They demand both unquestioning acceptance of nonsense such as 'transwomen are women', '(anthropogenic) climate change is real', while pretending to deny any objectives means for deciding what truth is, and at the extreme, even proclaiming truth to be nothing more than white racism.
Expertise, and 'Argument from Authority' are Authoritarian, and Elitist.
Many Academics want to be able to dictate how we think the world works by "appeal to expertise", or appeal to authority. It follows that the biggest experts, in their view are academics with doctorates. To achieve their objectives many academics are happy to throw science under the bus. Not just science, but free-speech, open debate, and freedom in general are often trashed by these leftist authoritarians. Today, 'fact-checkers' are censors. They are 21st century bureaucratic censors. Empowered with their science studies, and media studies degrees to dictate what the public are allowed to say and think. Not just the public, they are now censoring actual scientists too. Singham's ideas are really a way to give armchair experts authority. Such 'experts' often don't do empirical work such as tests, experiments, observations, validations & falsifications. They do easy-chair work such as modelling, statistics & theory. Theorists, modellers, and fake statisticians are contemptuous of those who get their hands dirty. People like Tesla, with his hundreds of patents and lack of degrees, are scum to such academics. Richard Feynman is somewhat hated by them too. Empiricists such as Tesla and Feynman understood the philosophy of science is really very simple. In contrast armchair theorists like Singham believe only experts with doctorates in science studies should be allowed an opinion - not just on what science is, but also on how to do science. To make that happen they're happy to go along with censorship, every step of the way.
