[lisper]

Ah.

FYI, I have a whole series of unpublished chapters about the (very messy) history of science. I decided to take this different approach because I thought it would be a more effective way of reaching my target audience.

> Science as a human enterprise has flourished best without anyone putting boxes on what does and what doesn't consitute science.

That's not true. There is one box you cannot get out of without destroying the process: any hypothesis that is inconsistent with experiment must be rejected (and as a corollary to that, unfalsifiable hypotheses must be rejected). But yeah, beyond that pretty much anything goes.

[AnimalMuppet]

I'm with you that, if you don't close the loop via experiment, you don't have science. But you have to be careful that you not go too far, and say that science is the only way we can know truth. Logical positivism is dead for a reason. (Actually, for several reasons.)

[lisper]

> science is the only way we can know truth

You need to read the third installment in my series:

https://blog.rongarret.info/2024/04/three-myths-about-scient...

and in particular Myth #3.

[jfengel]

I don't disagree with Myth #3, but I had a hard time getting to it after reading Myth #2:

The scientific method assumes naturalism/materialism/atheism... This is false. The scientific method contains no assumptions whatsoever. The scientific method is simply that: a method.

I don't think that's a good way to put it. There is an assumption that this method is a valid, useful, good thing to do. A souffle recipe presupposes that you want a souffle, or you wouldn't be reading it.

The scientific method hints ambiguously at epistemological commitments, and that ambiguity is not a point in its favor. Different people make those assumptions tacitly and don't realize that they disagree with each other, and don't even apply them consistently to themselves.

I also happen to agree that the scientific method has some kind of epistemic benefit, especially as compared to the potential alternatives. But those benefits prove maddeningly difficult to nail down. The epistemic commitments always turn out to be too loose (admitting pseudosciences) or too strict (rejecting sciences that resist the kinds of experiments you'd like to do).

[autoexec]

> I don't disagree with Myth #3,

You could. It says "First, science never proves anything; instead it produces explanations of observations." but science doesn't have to explain an observation. There are tons of times when science tells us what we'll observe under certain conditions while never giving us an explanation for how/why it works out that way. An explanation is the ideal, but all science needs is to give us something we can reasonably predict to be useful.

[lisper]

> There is an assumption that this method is a valid, useful, good thing to do.

No, that is not an assumption. That is an observation. The scientific method produces theories with predictive power, and it does this better than any other known method. That is an empirical fact, not an assumption. This is the reason science is a thing.

[jfengel]

Science has not proven that the observation will continue to hold. It happens to have worked so far, but you cannot prove that it will continue to.

You're applying the scientific method to itself, making a large-scale inference from limited data. This is precisely the grounds on which the Logical Positivists realized that they couldn't achieve their ends of putting science on a sound scientific footing.

[mikhailfranco]

All facts are provisional. If a validating observation does not continue to hold, that is another observational fact. At some point of replication and sigmas, it becomes another Problem, to be fed back into the scientific meat grinder.

For example, maybe vague observations of Mercury were consistent with Newton. But eventually, telescopes improved, duration of accurate records extended, many good instruments were deployed around the world. Then it became a Problem. Also note that Einstein fitted GR to Mercury, so he could not predict it as independent confirmation. It took the Eddington observation of light bending near the sun, during a solar eclipse, to provide the first evidence (even that was weak, and fixed, but the fix was on the right side of history :)

https://en.wikipedia.org/wiki/Eddington_experiment

[lisper]

> Science has not proven that the observation will continue to hold. It happens to have worked so far, but you cannot prove that it will continue to.

Yes, that's true. So?

> You're applying the scientific method to itself

Yes, that's true too.

> This is precisely the grounds on which the Logical Positivists realized that they couldn't achieve their ends of putting science on a sound scientific footing.

Maybe, but I'm not making any claims about knowledge. The only claim I'm making is about effectiveness and that's an empirical claim. not a philosophical one. One can go on to hypothesize that science is effective because it is "true" (whatever that means). That seems plausible to me (for some reasonable definition of "true") but it's still just another hypothesis.

[6510]

I'd fill that last bit under unnecessary and undesirable processing. Things are what they are(!) A hypothesis inconsistent with experiments is a hypothesis inconsistent with experiments. It is a mistake to attempt to make more or less from it. You get nothing extra or beneficial out of the effort of rejection. A failed experiment is a failed experiment. There are lots of ways to fail. Faulty materials, defective equipment, outside influences, incompetent experimenters, poor design, POLITICS etc etc

A repeatedly failed experiment sounds just as bad as it should.

[lisper]

> repeatedly failed experiment

That is an oxymoron. A repeatedly "failed" experiment is a valid scientific result. There is even a classic example:

https://en.wikipedia.org/wiki/Michelson%E2%80%93Morley_exper...

[fastasucan]

>hypothesis that is inconsistent with experiment must be rejected

Not really, this is a conclusion that is drawn only by looking at the big names and highlights throughout the scientific history. In practice the experiments in themselves is flawed, so if the hypothesis is not confirmed through experiments the experiment design need to be fixed. Other times they are confirmed where the experiments also was flawed.

Also there are a lot of hypothesis we cannot reject, but we still build scientific work upon them.

[lisper]

> In practice the experiments in themselves is flawed

Yes, of course. Experimental error is always a possibilities, but that is just another hypothesis that needs to be considered along with all the other possibilities.

> Also there are a lot of hypothesis we cannot reject, but we still build scientific work upon them.

Like what?

[raskelll]

Let me revisit my statement once I've read those chapters then.

Regarding your second statement, out of curiosity, I wonder how you would characterize the 1989 series of experiments done by Fleischmann and Pons.

[lisper]

> Let me revisit my statement once I've read those chapters then.

You can find them here:

https://flownet.com/ron/TIKN/

Note that they are quite drafty. That was the result of a previous effort to tackle this project of writing about the scientific method for a general audience that I ultimately abandoned in favor of this current approach. The history part starts in chapter 5.

These were based on a series of lectures I gave a few years ago. I can dig up those links too if you're interested.

> I wonder how you would characterize the 1989 series of experiments done by Fleischmann and Pons.

I'm not sure what you expect me to say. The results have not been reproduced, so whatever happened in 1989 it was almost certainly not cold fusion.

[vixen99]

No, we don't yet know what's happening in that area: as the DARPA 2022 report puts it in a press release entitled 'Solid State verification of nuclear particles in electrochemical cells'. "Work should continue - much interesting science to be done. Results do not yet rise to level publishable in peer-reviewed physics journals". Bottom line: some neutrons appear and they cannot explain the mechanism. If neutrons are detected and it's definitely 'cold' then what would you call it?

There are many other reports more or less with the same proviso. 'Not yet publishable'. Many will wonder about the 'yet' but that's usually the case with potentially new areas of science.

https://doi.org/10.1063/5.0153487 Li–Pd–Rh-D2O electrochemistry experiments at elevated voltage

[lisper]

I would call it "anomalous room-temperature neutron production".

[maxnoe]

This is easily solved by adding the well known quote (Sagan):

"Extraordinary claims require extraordinary evidence."

An older quote, addressing this case more directly would be (Eddington):

"The law that entropy always increases holds, I think, the supreme position among the laws of Nature. If someone points out to you that your pet theory of the universe is in disagreement with Maxwell's equations — then so much the worse for Maxwell's equations. If it is found to be contradicted by observation — *well, these experimentalists do bungle things sometimes.* But if your theory is found to be against the second law of thermodynamics I can give you no hope; there is nothing for it but to collapse in deepest humiliation."

[lisper]

Well, Fleischmann and Pons claimed to have extraordinary evidence. They were just wrong.

Also, I cover the Sagan quote in the previous installment:

https://blog.rongarret.info/2024/04/the-scientific-method-pa...

[maxnoe]

You don't get to claim yourself that the evidence you obtained is extraordinary

[mistermann]

Go through Lisper's history, he is quite literally never incorrect, even when he has no evidence or outright dodges all questions.

His performance in a philosophy thread a while back was extraordinary, explaining the truth and true value of various philosophy and philosophers, despite not having read it.

This is the power of The Science, nothing is more powerful, and nothing can be more powerful.

[mikhailfranco]

Yes, extraordinary is just a function of replication (number, diversity, quality, reputation) and sigmas. It has to satisfy a significant majority of scientists (peers), with diverse relevant qualifications (theoretical, experimental, across relevant disciplines), for a significant time.

[lisper]

They didn't, and that's obviously not what I meant.

[mikhailfranco]

No, the 2nd Law is just a statistical property relating macrostates to microstates. It can never be the foundation of physics for the underlying system.

Life does a pretty good job of evading it, locally, for lifetimes. Civilizations overcome it for longer, perhaps indefinitely, as long as there are stars/blackholes in the sky, or uranium/hydrogen nuclear fuel to be scavenged.

Wolfram recently tried to explain the 2nd Law, as a consequence of time-coherent computationally-bounded observers. As always with him, it is fascinating and infuriatingly in equal measure:

https://www.wolfram-media.com/products/the-second-law-resolv...

Those subsystems that evade it (life) also need explanation. Friston, Levin, Lane, England and others are starting to give plausible models and explanations.