Falsifiability -- What is Popper proposing and how should it read?

[Julius Ceasar]

TL;DR Summary

From Wikipedia, the free encyclopedia

I see falsification depends on a demonstrated false observation
wiki
(A statement, hypothesis, or theory is falsifiable if it can be demonstrated to be false by observation.)

But the following seems to imply that unless a statement is demonstrated false it is unscientific.

wiki
(The concept was introduced by the philosopher of science Karl Popper. He saw falsifiability as the logical part and the cornerstone of his scientific epistemology, which sets the limits of scientific inquiry. He proposed that statements and theories that are not falsifiable are unscientific. Declaring an unfalsifiable theory to be scientific would then be pseudoscience.)

I don't think this is what Popper meant but this is----- how it reads, What is he proposing and how should it read?
Perhaps he proposed that statements and theories must apply rigorous and skeptical observations to be truly scientific but there is always the problem of induction that some observation has not been accounted for, ala 'all swans are white', then finding black swans.

I see the benefit of trying to falsify statements and theories but it is pretty useless if you can't check everything.

 

[mfb]

Your confuse "false" and "falsifiable".

To be scientific it must be possible to show that a statement is false.
To be relevant this must never have happened.

"It will rain here tomorrow" is falsifiable: We can wait until tomorrow and check.
"It rained here yesterday" is falsifiable and false: It didn't rain here yesterday.
"There is an invisible unicorn that doesn't interact with the world in any way" is not falsifiable. There is no way we could check this statement. No matter what you observe, it cannot be in contradiction with the statement. Even if you find a visible unicorn: That doesn't rule out the existence of the invisible one.

 

[Julius Ceasar]

Thanks mfb that does make sense, it must not be impossible to check.

 

[Ibix]

Right. Another way of saying it is that, to be scientific, a hypothesis must make testable predictions, and that there must be some outcomes that are (at least in principle) possible and are not consistent with the hypothesis. We know from experience that there are days that it rains and days that it doesn't, so mfb's "rain tomorrow" hypothesis remains scientific even if he's got an iron clad weather model that has passed 100% of tests to date.

 

[Julius Ceasar]

Summary:: From Wikipedia, the free encyclopedia

He proposed that statements and theories that are not falsifiable are unscientific

He proposed that for a statement to be classed as scientific it must have a possible outcome that can prove it wrong. I don't agree with Popper here, QM for instance becomes unscientific because there is no outcome to prove duality wrong. Also the simplicity of some statements leave no room except a single outcome like, 'it's always now'.

I did understand what popper was saying but wanted to point out how language is sometimes a barrier and can even make it seem like the opposite meaning.

 

[Nugatory]

QM for instance becomes unscientific because there is no outcome to prove duality wrong

This argument doesn’t work because duality is not a prediction of quantum mechanics. The idea was abandoned close to a century ago and appears nowhere in the modern (that is, after about 1930) understanding of the theory. Unfortunately, like consciousness observers and Schrodinger’s cat that’s neither dead nor alive until we open the box and look, duality has taken root in the public imagination and lives on as a sort of urban legend.

 

[Julius Ceasar]

OK i didn't realize duality wasn't a part of QM anymore, what was the result?

1.particles are also waves
2. particles are only particles
Both statements are falsifiable, which statement was wrong?

 

[DaveC426913]

Hasn't duality become part of the interpretation of QM, which is why it's not subject to, er ... falsificative rigor?

 

[mfb]

OK i didn't realize duality wasn't a part of QM anymore, what was the result?

1.particles are also waves
2. particles are only particles
Both statements are falsifiable, which statement was wrong?

Both.
Things are quantum objects, they are neither particles nor waves. They have some properties similar to classical particles, they have some properties similar to classical waves, but they are neither.

 

[Julius Ceasar]

So what are quantum objects, can you make a falsifiable statement to describe them?

 

[Ibix]

I don't think you need to be able to falsify "quantum objects" in the abstract like that. It's just a definition.

However, the claim "the electron is a quantum object", or "all particles are quantum objects" are scientific claims because we are now saying that some real-world object behaves according to that definition. Such claims are subject to falsification. For example we are now claiming that you can get a diffraction pattern from a repeated single particle two slit experiment using electrons (or whatever). Getting a diffraction pattern, not a single dot, from one run of the experiment would falsify the claim. So would repeating the experiment and getting a pattern that was a shadow of the slits, not a diffraction pattern (in a regime where those two things are predicted to be distinct).

 

[Julius Ceasar]

You are definitely saying two opposite things here by saying 'however'

"the electron is a quantum object"

So is this falsifiable?

To be scientific it must be possible to show that a statement is false.

It is falsifiable because the DSE shows the statement false, i.e. an electron is also a classical object when observed.
I am trying to demonstrate falsification is not all Popper hoped it would be.
there are other examples.

edit- not your claim i think

 

[Ibix]

I'm not saying opposite things - just different things. "A unicorn is a horse with a horn on its head" isn't a testable claim - just a definition of the word "unicorn". It's independent of the existence or not of unicorns. I can make scientific claims about unicorns (which I do not expect to be true), but the mere definition is not a scientific claim.

Similarly "a quantum object has some of the properties of a classical particle and some of the properties of a classical wave" is a mere definition. There isn't anything to test unless I assert that some real world object matches that definition. For example, an electron.

Yes, the claim that an electron is a quantum object is a falsifiable claim. I already stated two possible outcomes of a double slit experiment that would be inconsistent with the claim. I don't understand what you mean by "an electron is also a classical object when observed".

 

[Julius Ceasar]

I'm not saying opposite things - just different things. "A unicorn is a horse with a horn on its head" isn't a testable claim - just a definition of the word "unicorn". It's independent of the existence or not of unicorns. I can make scientific claims about unicorns, but the mere definition is not a scientific claim.

understood

Similarly "a quantum object has some of the properties of a classical particle and some of the properties of a classical wave" is a mere definition. There isn't anything to test unless I assert that some real world object matches that definition. For example, an electron.

However, for example, if you assert, you can make it testable, kool i get, all good.

Yes, the claim that an electron is a quantum object is a falsifiable claim. I already stated two possible outcomes of a double slit experiment that would be inconsistent with the claim. I don't understand what you mean by "an electron is also a classical object when observed".

I was simply agreeing with your previous statement, electrons act like little balls of matter when observed, i.e. a classical object.

 

[Julius Ceasar]

However, the claim "the electron is a quantum object", or "all particles are quantum objects" are scientific claims because we are now saying that some real-world object behaves according to that definition. Such claims are subject to falsification.

So the DSE shows the electron is not a quantum object because (when tested) it acts like a classical object when observed. Right? But in its natural unobserved state it is a quantum object. Right? Then Poppers assertion that it's not scientific without falsification seems useless in this instance and perhaps others.

 

[Ibix]

I was simply agreeing with your previous statement, electrons act like little balls of matter

I see - you mean they behave like a classical particle, I think, rather than a classical wave. The relevant point, though, is that the prediction is that you will find these "classical particles" in places where they couldn't possibly be if they actually were classical particles.

Actually doing the experiment immediately falsifies the claim "electrons are classical particles" because you find them where they couldn't be if that claim were true. The claim "electrons are quantum objects" is then our current best model - we've never seen them behave inconsistently with the predictions of that model.

 

[Ibix]

So the DSE shows the electron is not a quantum object because (when tested) it acts like a classical object when observed. Right?

No - because the electrons show up in places they couldn't reach if they were classical particles. Furthermore they show up in the places predicted by our quantum model in exactly the quantities predicted, to the limits of our measurement precision.

 

[Julius Ceasar]

Yes and no...'sometimes' is the point. falsification does not work in this instance. Stating some evidence while ignoring others is bias.

 

[Ibix]

To put it another way, the claim "electrons are classical waves" predicts that any double slit experiment will give you a smooth diffraction pattern (edit: a very specific diffraction pattern with a clearly specified structure). The claim "electrons are classical particles" predicts that a high intensity electron beam will make a shadow of the two slits, and a low intensity beam will give you individual dots that build up over time into the same shadow. The claim "electrons are quantum objects" predicts a high intensity beam will give a diffraction pattern (edit: again, a specific, quantifiable pattern) and a low intensity beam will give individual dots that will build up over time into the same diffraction pattern.

There are three completely distinct predictions here. Getting dots at low intensity falsifies "classical wave". Getting a diffraction pattern at high intensity or over time at low intensity falsifies "classical particle". (Edit: and getting the wrong pattern, or a smooth diffraction pattern at low intensity would falsify "quantum objects", but that doesn't happen in practice.)

Ignoring the patterns in the dots would be, as you posted while I was typing this, bias.

 

[mfb]

Yes and no...'sometimes' is the point. falsification does not work in this instance. Stating some evidence while ignoring others is bias.

No, "sometimes" is a one-word summary of what you can learn over several years. Quantum mechanics makes very precise predictions how electrons behave when. A one-sentence description in English can never catch all the predictions fully. We can (and did) test these predictions, and they match experiments. Every. Single. Time.

 

[Julius Ceasar]

Yet its all to do with one statement

 

[Nugatory]

So what are quantum objects, can you make a falsifiable statement to describe them?

Statements of the form “If we measure X the probability of result Y will be Z” are falsifiable - and these are the only statements that quantum mechanics ever makes.

 

[OCR]

Statements of the of the form. . .

.

 

[Nugatory]

View attachment 254986

.

Yes
(Thank you, I fixed it)

 

[OCR]

(Thank you, I fixed it)

✔ . . ☺

.

 

[Julius Ceasar]

I have another question in this old post after spending some time in a post called,"On the relation between physics and philosophy".
Is the claim, "all swans are white and have always been white" the same as the question, "do only white swans exist", which appears philosophical? One is a claim that could be proved false and the other is a question that could be proved false, is this sentence preparation all it takes for philosophy to become physics?

 

[Ibix]

Both claims seem falsifiable to me, and routinely are falsified by the existence of black swans. Although one could interpret the latter as part of a definition of a swan (i.e., that it is part of a statement that anything that looks like a swan but isn't white is correctly referred to as a not-a-swan), in which case there's an argument about semantics to be had before you can make a decision about falsifiability.

Ultimately, if the two statements mean the same thing, then they must both be testable claims or both not be. I would suspect that there might be argument about whether they do mean the same thing.