Yes, the hypothesis that Neptune exists (A) would make the surprising anomalies in the orbit of Uranus (C) a matter of course; therefore, we have reason to suspect that Neptune exists. — aletheist

And by the same logic, we have reason to believe the statement "Vulcan exists", is true?

Scientific hypotheses are causal arguments. Causal arguments are inductive. — TheMadFool

Scientific theories can't be causal arguments if they work just as well backwards in time.

That the Earth is spherical is a scientific hypothesis. There were two competing hypotheses:

1. Earth is flat
2. Earth is spherical — TheMadFool

They are empirical statements, but "it is raining" is also an empirical statement, and no one claims that is a scientific theory.

Popper argued that scientific theories take the logical form of "strictly universal statements", and none of the "singular statements" above fall into that category.

Big shout out to Eratosthenes.

1. Identify problem (observational findings) — TheMadFool

Sometimes "observational findings", but also purely theoretical considerations. For example the huge and varied research programs to unify quantum mechanics and general relativity have nothing to do with anything observed. And where it is "observational findings" it will be due to prevailing theories being put to the test. Physicists don't just stumble across gravitational waves.

For example, the problem that Special Relativity sought to solve, was the unification of electromagnetism and Newtonian mechanics.

2. Construct hypothesis — TheMadFool

You make it sound so easy! It is worth noting, that prior to Popper, the Logical Positivists admitted theories to science once they were verified. For Popper, theories are automatically admitted, for only then can they be criticised, tested, and falsified.

3. Test hypothesis. This involves making and checking predictions — TheMadFool

You could make predictions using Newton's laws and check them all day long and you won't encounter any surprises. The crucial point is error elimination, and testing is just part of your arsenal.

4. Revise hypothesis if hypothesis fails to account for all observations — TheMadFool

Again, the problems discovered are often nothing to do with observation, though they can be. And you might need a whole new set of ideas.

5. Identify further implications — TheMadFool

And if you're lucky, you might get to refute your cherished ideas.

However, I'm specifically talking about step 3 of the process -testing the hypothesis. In this case we do commit the logical fallacy of affirming the consequent as below:

1. If hypothesis A is true then predictions B, C, D,...are true
2. Predictions B, C, D,... are true
So,
3. Hypothesis A is true

What are your comments on this? — TheMadFool

No one is claiming A is true, or even B,C,D are true. But if you claim that B,C,D are false, and you have theories why that is the case, then prospects for A are looking bleak.

Peirce's favorite example was Kepler's series of hypotheses regarding the orbit of Mars. After only a few unsuccessful conjectures, he tried an ellipse, which not only fit the data that he had from previous observations, but also led to predictions that were subsequently corroborated by further observations. — aletheist

Well, trial and error can certainly work when trying to figure out the shape of orbits, but lets have a more recent example.

You are doubtless aware that the orbit of Uranus did not seem to agree with Newton's law of gravitation, so we have:

C=Anomalous orbit of Uranus.
A=Neptune is retrodicted to exist, which would make C "unsurprising".
Therefore we have "reason" to believe "Neptune exists" is true.

Is that a fair representation of Pierce's method?

That is to say it is logically impossible for Popper's criterion of demarcation to be false? Or do you mean to say that because it plays a prescriptive role it does not make sense to say his criterion is true or false? Or what? — Moliere

It is a matter of judgement whether a statement is falsifiable or not. Some people think Freudian psychoanalysis is non-falsifiable, others disagree.

Despite what is claimed for the principle of demarcation, and I am as guilty as anyone, it is not really about deciding what is science and what is not. Such questions can be left to the institutions. Rather, in Popper's words, it is to solve,

...an urgent practical problem: under what conditions is a critical appeal to experience possible - one that could bear some fruit?
(Realism and the Aim of Science, 1983)

So really it is about what sort of statements are amenable to empirical investigation. These are the falsifiable statements.

And I think it logically impossible for true ideas to be false.

Popper's position, and mine, is that when a hypothesis graduates to a higher state it becomes a theory. There is no higher state than a theory. To say that a theory is not a fact is not to question it, it is to acknowledge the proper accepted meaning of the word theory in the scientific community. — andrewk

Popper does not differentiate between an hypothesis and a theory. The only way conjectures can be dealt with is to admit them to the scientific method.

A fact is a raw observation — andrewk

No such thing. All observations are theory-laden.

A well-prepared mind. Again, Peirce called it a conjecture long before Popper did. — aletheist

OK, so accepting, for the sake of argument, that given a scientific problem, we manage to retrodict a solution, what are the "reasons to suspect" that A is true. Can you give an example?

Charles Sanders Peirce spelled out the scientific method (as outlined here) the year after Popper was born. — aletheist

Where does "A" come from?

1) If hypothesis A is true then predictions B, C, D, etc. are true
2. Predictions B, C, D, etc. are true
Therefore
3) Hypothesis A is true

Is the above the correct blueprint for all scientific arguments? — TheMadFool

Nay, Nay, and thrice Nay!

The Scientific Method goes like this:

1) Problem Identification - You find a problem you are interested in or need to solve.

2) Tentative Solutions - You conjecture solutions to your problem.

3) Error elimination through criticism and testing of your conjectures, refutation.

4) Further conjectures.

5) Further refutations, and so on without end.

Within Popper's method, it needs to be emphasised that nothing is conclusive. Falsification happens, but conclusive falsification cannot happen. Also, contrary to the Logical Positivists whose ideas were dominant at the time Popper wrote LSD, theories are not verified, and thus admitted to science. Theories are automatically admitted, but if they fail, they are rejected.

Let's not forget that Popper also rescued metaphysics with his method.

We can't directly observe that a planet, or any object, is spherical because we see in 2D and spheres are 3D. We have reams of data that are consistent with the theory that the Earth is approximately spherical so we adopt that theory.

I set the boundary between observations and theories much lower down - towards the very raw input end. — andrewk

And yet Eratosthenes was able to measure the circumference of the Earth in ~200 B.C.

The scientific method deals with universal statements, and I'm not convinced that deducing the shape and size of a particular object (albeit a large and rather important one) counts as one of those.

But there are theories, which are undeniably scientific theories, and are undeniably true, for example plate-tectonics, the age of the earth, even dinosaurs. These theories seem to go from being a controversial hypothesis to a fact of reality, and anyone who questions them is a crackpot.

In defence of Popper in such situations, we need to remember that Popper never claimed that scientific truth could not be achieved, rather than it could not be certified. And the logical status of these theories is that no viable criticism of them exists. Are you willing to commit to the statement that no criticism of them can ever exist?

Evolution, or rather the "Modern Evolutionary Synthesis", does seem to be the elephant in the room in this regard, though. A core theory to our conception of reality, that, contra-Popper, is certifiably true, it seems.

Popper, who is deservedly admired, set out a logic for science that is pretty neat. But the problem with logic, as Feyerabend showed, is that people are only occasionally logical. — Banno

No he didn't. Popper argued that it is logically impossible to verify or falsify a scientific theory. He sets out a Method, based on certain epistemological truths he discovered. That's why it's called the Scientific Method.

Nevertheless, there is an asymmetry between verification and falsification when it comes to universal statements, and this is what he exploits in his method.

But you can answer the question too. Suppose the criterion of demarcation does not hold, and not only that that there is no such criteria. What would be the more honest approach? To invent a more elaborate theory of science with better fidelity, or to point out that there is no such theory? — Moliere

Before Popper's criterion of demarcation between science and non-science, the prevailing wisdom was that of the Vienna Circle. The Logical Positivists distinguished between science and nonsense via their criterion of meaning.

Popper not only demarcated science this way, but in the process rescued metaphysics.

There is no sense in which Popper's criterion of demarcation "does not hold". You can only be on one side of it or the other, though you can be mistaken about which side you think you are on.

My problem is, how does this relate to the theory of a spherical Earth? It seems here we have a theory that survived falsification and has gone on to become fact through confirmation via observing the spherical shape of the Earth thanks to space exploration. — Craig

There is no "theory of a spherical earth". The earth literally is (approximately) spherical. The theory is the explanation of that phenomenon, and there have been a couple of those.

But sure, if you go far enough back in time, perhaps you will find the mistaken conclusion that the Earth is flat, but that misconception was the result of inadequate theory, and not itself a theory. In Popperian language, the flatness of the Earth would be a basic statement, that could be (in principle) tested.

Does the spherical Earth cast doubt upon Popper’s claims about scientific theories never been confirmed? — Craig

You've got to be a bit careful here. Wasn't it Hume who first noticed this, and then Kant who wrestled with the problem for a large part of his career?

You will be surprised to learn the Popper not only argued that scientific theories cannot be verified, but that they cannot be logically falsified either, before going on to solve the problems set up by Hume.

It could be that the Earth we see as approximately spherical is actually a 3D cross-section of an object that is actually a 4D hypersphere. We can never rule out more elaborate theories. But we don't need to. We just say 'this is the best hypothesis we have at present, and it has been working very well, so we'll keep on using that unless or until it stops working well'. — andrewk

That the Earth is spherical is not a scientific theory, it is merely a fact of reality.

The scientific theory consists of the explanation for this and associated facts.

Nonetheless, Popper still deserves credit for reinvigorating old ideas. — Pelle

He really didn't just invigorate old ideas. They were his ideas and they were so radical, that still very few understand him.

Science is about solving problems, but Marxism nor Creationism solve problems: they merely provide explanations taken directly from their ideological framework (which is essentially a set of conclusions). — Pelle

And as a true Popperian you reject authoritative sources of knowledge, certainty, and anything that claims to beyond criticism.

What if there is no specific set of criteria that captures all science? Wouldn't it be more honest to not describe it if that were the case? — Moliere

There is: The Criterion of Demarcation.

I guess the response here is -- so what? If someone wants to run a research program on Marxist Science, Creation Science, and Astrology, who cares? I can tell you the specific reasons why I don't believe in this or that set of beliefs. But there's no reason to have an over-arching theory of knowledge to safeguard the sanctity of academia. I can respond in kind to any sort of research program or argument. — Moliere

This would be against the Scientific Method, and ruled out by the criterion of demarcation.

modern science does follow Popper's ideas to some extent. The critical discussion around science today is exactly as Popper described: people trying to falsify eachother's theories. — Pelle

I'm not so sure this is an accurate characterisation of science, or quite captures Popper's central ideas. Let's not forget that Freudian psychoanalysis still exists, as does the Copenhagen interpretation. Also, the Sokal affair, and the more recent Sokal-Squared debacle in the Social "sciences".

I suspect Popper might have some advice for String theorists too. His method is much more iterative, focusing on identifying specific problems that need to be solved, rather than the all-or-nothing grand-plan approach of their research program. Ironically, the earliest developments in String theory came about by focusing on specific problems.

It is interesting to note, that the floundering efforts to unify physics may be rescued by an avowed Popperian. See the recent papers by Chiara Marletto
https://arxiv.org/search/quant-ph?searchtype=author&query=Marletto%2C+C

Popper's account of science is much more akin to the struggle to find good realist accounts of what exists in reality, how it behaves, and why it does so.

completely agree. The biological aspects of Popper's writing is what I found the most interesting. There's seems to be a lot of people that think he's a status quo shill with no radical ideas, but it's just not true. — Pelle

Couldn't be further from the truth.

Popper will be underestimated for a thousand years.

I'll wait on you providing the reference later, then. — fdrake

I tell you what, in the absence of a block function on this forum, please never interact with me again, and I will pay you the same courtesy.

LSD section 19: "...the theoretical systems of the natural sciences are not verifiable, but I assert that they are not falsifiable either.

LSD section 6: "...it is still impossible, for various reasons, that any theoretical system should ever be conclusively falsified.

Ok! — fdrake

It's literally in the book, and I would provide an exact quote but there is a 4 year old sleeping in the room where the book is. Look it up!

They do! The only problem is that they do not resemble how science is actually done in a universal sense. — Moliere

It's exactly how all new theories are treated in Science. There is no other way to the truth.

Firstly, Lakatos characterises falsification as operative not on singular propositions, but on series of propositions. Such a series might be 'The laws of Newtonian mechanics + observations about Uranus' orbit' or 'The laws of Newtonian mechanics + observations about Mercury's orbit', and when falsification strikes (when 'Nature shouts "No!" as he puts it), it does not act on a specific proposition, but on the composites "The laws of Newtonian mechanics + observations about Uranus' orbit'. — fdrake

Popper confronts and solve the Duhem-Quine Thesis in "Logic of Scientific Discovery".

The second difference from Popperian falsification is that the rejection entailed by Nature shouting 'No!' is weakened. We don't reject Newtonian mechanics entirely just because it fails to model the orbit of Mercury, we rather constrain its application to a domain of relevance, and this is done adaptively with respect to theoretical and experimental demarcations. — fdrake

According to Popper, falsification of any theory is logically impossible.

More philosophically -- the problem with Popper is the scope of his claims, and the prescriptive nature of his project. He's making a normative project for scientists, and doing so not just for a few scientists who feel inspired but for scientific knowledge as a whole. And while Popper's method gets at some aspects of scientific thought, it does not meet the burden it sets for itself -- and yet still demands that science should be performed in accordance with his particular epistemological concerns. — Moliere

What sort of prescriptive things do you not like? The requirement to be open to criticism, and to subject one's ideas to the harshest of tests? How a bout the requirement to not appeal to any authority, or that a scientific theory must yield testable statements about reality?

There are more subtle requirements, like given the choice of two theories, one should choose the one with higher empirical content, because it will be more falsifiable. Also, that falsification is itself subject to error (we are fallibilists after all) so any such decision must be regarded as tentative, so theories and their criticisms are never completely discarded.

Of course, this is only a superficial account, but these ideas seem rather good to me.

Complaining about the Scientific Method is like complaining about Evolution.

Feyerabend demonstrates this by placing Popper's method alongside Galileo's Dialogue Concerning the Two Chief World Systems -- an example which surely everyone would agree is properly scientific, and even good science, yet does not follow Popper's method. So either Galileo is wrong about how to do science, or Popper has overstated the scope to which his method applies. (at least, if we agree with Feyerabend's analysis, of course -- we could set out to save Popper by trying to reframe Galileo in Popperian terms. But I'm fairly well convinced by the arguments in Against Method) — Moliere

As is common in most misunderstandings of Popper and the Scientific Method, is the misconception that Popper provides rules for solving problems or constructing theories. He does not, because there is no such thing. Popper's Scientific Method deals solely with how we must treat and criticise theories, not how we come by them.

NOTA is what the Monster Raving Loony Party is for.

In the field of epistemology, there are three special actors who will never cease to influence: Thomas Kuhn, Paul Feyerabend and Karl Popper. Each brought their respective doctrines; subjectivist scientism (crassly put), epistemic anarchism and falsificationist realism. These epistemological concepts all painted science in a different light and came in turn to fight bloody battles in the court of ideas for many years to come. — Pelle

I wince a little when Popper is called a "falsificationist ". Not because he was not a falsificationist, but because falsificationism is so often misrepresented as naive falsificationism. Popper did after all, in LSD, explain that logical falsification of a scientific theory is impossible. He was however, certainly a "Critical Rationalist" of which "fallibilism" is a central tenet.

What were Popper's main achievements in epistemology? Wasn't he first to elucidate the scientific method? He rehabilitated metaphysics from the Logical Positivists, gave us the criterion of demarcation, and solved the problem of induction. Plenty of other achievements, I'm sure, but I would need to consult a book to compile the list.

It might be worth mentioning that Popper was heavily involved in the intellectual circle that developed the Modern Evolutionary Synthesis, where his epistemology and how genetics works are strikingly similar.

Partly because of the above associations, several Nobel Prize winners in the fields of biology and medicine credit their discoveries to following Popper's method. His ideas have also been central in discoveries in physics and computing. I'm not sure the other epistemologists mentioned could make a similar claim. Also Popper's ideas still seem quite active, in the sense that there appears to have been some progress recently.

One thing is for sure, Popper must be the most misrepresented and most misunderstood philosopher of the 20th century.

Then the question is: how do we understand what [the other thing] is? — csalisbury

The same way we understand what anything is. We encounter a phenomenon of some kind, perhaps a feature or a regularity, and we conjecture an explanation for it. etc.

Now you have to explain how matter produces an understanding of something other than matter. — csalisbury

We still need to figure out how matter understands matter though. And since "understanding" is not matter, that seems quite a tricky problem in itself.

It's not a nonsequitur, because then logic is "above" god so to speak. — Terrapin Station

If your conception of God is an inconsistent, incoherent, self-contradictory, unnecessary entity, then I'm not sure the point of engaging with the idea.

That doesn't have the same dilemma built into it though. The "rock heavier than he can lift" thing sets up a dichotomy where either answer implies something a god wouldn't be able to do. — Terrapin Station

I think the difference is psychological rather than logical. We can all imagine things we can't lift, and even making things we can't lift, and this clouds our reasoning.

The paradox translates to something like this:

Can God create logical inconsistencies?
If not, then God is not omnipotent.

Non-sequitur if you ask me.

Either it's possible for him to create a stone that he can not lift or that's not possible. Both possibilities imply something he's not able to do (he either could not lift that rock that he could create, or he could not create such a rock). Whether he actually creates it or not is beside the logical point. — Terrapin Station

No one ever argues that god lacks omnipotence because she cannot make 2+2=5.

In short God is forced to not create the stone. Being forced to do/not do something implies that God isn't omnipotent. — TheMadFool

Since when does omnipotence require the ability to defy logic, to instantiate logical contradictions?

God can create a stone that he cannot lift, and if he does then he will lose his omnipotence, but if he doesn't then he remains omnipotent. — Michael

I don't think it reasonable to regard omnipotence as something you can lose just by making a stone.

No, I'm saying that God is able to, but because he doesn't, there isn't a stone that he cannot lift, and so he remains omnipotent.

Your mistake is in saying that if God can create the stone then there is a stone that he cannot lift. That doesn't follow. — Michael

I don't quite get the idea that an omnipotent/omniscient being is required to have logically impossible abilities, and have access to in principle unknowable knowledge. It seems we demand too much from our gods.

I don't see that being unable to make a stone so large he cannot pick up, or a rose that is simultaneously white and red, is a slight on his omnipotence?

What you are saying makes sense, but what is knowledge? What can you know surely?
I don't know if you are familiar with Rene Descartes but I based my questions mostly on his solutions. — Towers

Certainty in knowledge is impossible, worthless, and damaging. What you are seeking is an authority, to certify certain truths. There is no such thing.

But again, what is truth and what is better?
Besides, for scientific progress (which I believe you think as ”good”) a lot of things we generally don’t see as moral has been done. — Towers

If you value knowledge, then only certain moral systems will support that value. Are they objectively good? I'm not sure that question is answerable. Are they objectively better? They are, in exactly the same way that General Relativity is better than Aristotle's theory of gravity.

Ethics: ethical foundations have nothing to do with logic and ethical utterances are not true or false. — Terrapin Station

Is it not true that, in order to achieve scientific progress, and knowledge in general, that a certain morality is required? A society cannot progress towards the truth unless they value it, and are open to change and willing to subject their ideas to criticism.

If you value truth, then there are objectively better ways to act.

I think it's close but not quite because paganism also involves a kind of deism. — BrianW

It seems more like polytheism. What am I missing?

No one would talk about views in, say, quantum mechanical interpretations this way so doing so here frankly sounds stupid. — MindForged

And, you call me "stupid"?
— Inis
— SophistiCat

I have to say, you leave me little choice. — SophistiCat

If you have nothing new to add, perhaps seek a different thread. — Arkady

Yes indeed, people who can't comprehend Galileo's thought experiment, call me "stupid" and tell me to "seek a different thread".

Oh, to be in the presence of those with bigger brains than Galileo!

I feel like 'atheism' is the wrong word to use considering our inclinations to believe in supernatural (beyond the norm and unlike the norm) phenomena is and has been an intrinsic part of our thoughts and emotions because part of seeking to learn what we don't know is expecting to find that which we don't know. Perhaps 'supernaturalism' is a better fit. — BrianW

"Pagan" is another word.