Science thread

[rotting bones]

I think Raphael means that assuming hypotheses based on whether they are desirable or not rather than on whether they are true or not is fallacious. For instance, to use your example, always talking about the genetic component of crime because one is a racist who would rather crime be genetic rather than environmental due to that reinforcing one's racist beliefs indeed would be a red flag for being based on a fallacy.

All the interpretations of QM are compatible with the known facts.

[rotting bones]

I think you don't understand either "testing" or "interpretation".

There is no proof for pilot wave theory, no operational test which supports it over any other interpretation.

I don't say that any of the interpretations is wrong.* But they are all untested in the ways they differ from other interpreatations. The arguments in support of them are thus philosophical or subjective.

*Except hidden variable theory, but even there, there are qualifications— you can save the idea if you can swallow non-locality.

I don't think it makes sense to test the pilot wave theory. By construction, that would seem impossible. One time, a group did an experiment to construct a "philosophical argument' against Bohmian mechanics. But since it's compatible with the math, all they ended up saying that it feels to them that the materialist component of the theory is meaningless.

Since all the interpretations are compatible with the math, you can find physicists who hold all of them.

[zompist]

Since all the interpretations are compatible with the math, you can find physicists who hold all of them.

Yes, exactly. Which means pilot wave theory is fine as an interpretation, but it isn't more "tested" than the other interpretations.

[keenir]

If you want a more mainstream example, the Indian Civil Service is one of the most corrupt institutions in the world. In the qualifying exam, entrants are required to take a test on ethics, where they write long essays on the subject. This is the kind of solution you get when you replace systems analysis with values and symbols.

Writing a long essay is a sign of corruption?

Guys, nobody tell Rotting Bones about the Chinese Imperial Exam!

[Raphael]

If you want a more mainstream example, the Indian Civil Service is one of the most corrupt institutions in the world. In the qualifying exam, entrants are required to take a test on ethics, where they write long essays on the subject. This is the kind of solution you get when you replace systems analysis with values and symbols.

Writing a long essay is a sign of corruption?

I think that rotting bones' point is that if an institution has a massive corruption problem, making people in that institution write lengthy essays about the importance of ethics isn't going to solve that problem. And I think that, if someone else than rotting bones would have made the same point, you would have understood that point perfectly well.

[Ares Land]

If you want a more mainstream example, the Indian Civil Service is one of the most corrupt institutions in the world. In the qualifying exam, entrants are required to take a test on ethics, where they write long essays on the subject. This is the kind of solution you get when you replace systems analysis with values and symbols.

Writing a long essay is a sign of corruption?

Guys, nobody tell Rotting Bones about the Chinese Imperial Exam!

Europeans loved the Confucean examination system when they heard about it, and as a result a similar system for entry into the Civil Service turns up all over the world.

It's not a great system; though a huge step up from the previous approach, where officials outright bought their charges.

[bradrn]

I’m reluctant to get myself involved in this conversation, which started bad and hasn’t improved, yet here I go…

Most of the problems come from reifying the Schrödinger equation. It's not a physical thing and nonsense mulitiplies if you treat it as one.

A rundown of the strange things:

The Copenhagen interpretation: the "collapse of the wavefunction" and endless noodling over the concept of "measurement."
Feynman's view: everything is particles, but to calculate where you might measure them, you use a probability function that acts like a wave. No need for a collapse of anything, but you have to accept that the universe doesn't let you predict things, only predict probabilities.
Many worlds: nothing collapses because an entire new universe appears. About the extreme case of entia getting multiplicanda.
Pilot wave theory (Bohm): in addition to Schrödinger's there yet another, also unobserved wave.
Dark photon theory: I don't know the details except it posits an entirely new undetectable class of "dark photons".
Hidden variables: there is some unseen deterministic mechanism "below" QM. This one is actually experimentally disconfirmed.

I like Feynman's idea the best, as it really makes no assumptions or added entities. People who don't like it generally have a non-scientific attachment to determinism.

I’m not sure where you saw this ‘Feynman interpretation’? It’s not something I’ve heard of before, and I’m pretty sure it contradicts the mathematics. Wikipedia claims that Feynman never really discussed interpretations of quantum mechanics.

For myself, I feel quite strongly that the so-called ‘many-worlds’ interpretation is what corresponds most straightforwardly to the mathematics. It’s not helped by the badly misleading popularisations — despite the name, it is not the case that ‘an entire new universe appears’. Rather, it would be more accurate to say that when you make an observation, your own quantum state becomes entangled with the state of the experiment you’re measuring (‘decoherence’): the resulting state for the composite system of the experiment and you becomes a superposition of each state such that (you observe such-and-such an outcome)⊗(the experiment had that same outcome). The fact that those states are all superposed is what gives the ‘many worlds’ of the name, but they’re not ‘new universes’ in the sci-fi sense — they’re just different outcomes of which you can only experience a single one, because the probability of experiencing two outcomes at once is 0%.

[zompist]

I’m not sure where you saw this ‘Feynman interpretation’? It’s not something I’ve heard of before, and I’m pretty sure it contradicts the mathematics. Wikipedia claims that Feynman never really discussed interpretations of quantum mechanics.

Um, I saw it in Feynman. It's in QED and the third volume of his Lectures in Physics.

For myself, I feel quite strongly that the so-called ‘many-worlds’ interpretation is what corresponds most straightforwardly to the mathematics.

Not for me, because it ignores the final step of a QM analysis: squaring the amplitudes to get a probability. As I said, people start to think strange things when you start reifying the Schrödinger equation. Feynman's approach is to just not do that.

(When you get to quantum field theory, things get even stranger. If I understand it, the "particles" end up as standing waves in a horrifically multidimensional 'space'. Another reason not to get too certain about interpretations.)

[rotting bones]

Europeans loved the Confucean examination system when they heard about it, and as a result a similar system for entry into the Civil Service turns up all over the world.

It's not a great system; though a huge step up from the previous approach, where officials outright bought their charges.

Obviously, the answer is to introduce accountability, not ask the villains how well they know what goodness is.

[rotting bones]

As I said, people start to think strange things when you start reifying the Schrödinger equation. Feynman's approach is to just not do that.

Personally, I don't believe in math. I believe that math is like the legends on a map that can be used to keep track of what's really there. "What's really there" is matter by definition. I think letting the math tell us what's there is putting the cart before the horse. In the case of QM, the state of this "matter" can only be got at through further experiments, neither from the math like Feynman nor metaphysical consistency like Cockshott.

[rotting bones]

I think that rotting bones' point is that if an institution has a massive corruption problem, making people in that institution write lengthy essays about the importance of ethics isn't going to solve that problem. And I think that, if someone else than rotting bones would have made the same point, you would have understood that point perfectly well.

Thanks. Nice to know I'm not screaming into the ether.

[keenir]

If you want a more mainstream example, the Indian Civil Service is one of the most corrupt institutions in the world. In the qualifying exam, entrants are required to take a test on ethics, where they write long essays on the subject. This is the kind of solution you get when you replace systems analysis with values and symbols.

Writing a long essay is a sign of corruption?

I think that rotting bones' point is that if an institution has a massive corruption problem, making people in that institution write lengthy essays about the importance of ethics isn't going to solve that problem. And I think that, if someone else than rotting bones would have made the same point, you would have understood that point perfectly well.

I'm afraid that if anyone else had worded it like that, I'd still have been baffled.

thank you for clarifying.

[Travis B.]

I'm afraid that if anyone else had worded it like that, I'd still have been baffled.

thank you for clarifying.

I was baffled as well myself.

[zompist]

As I said, people start to think strange things when you start reifying the Schrödinger equation. Feynman's approach is to just not do that.

Personally, I don't believe in math. I believe that math is like the legends on a map that can be used to keep track of what's really there. "What's really there" is matter by definition. I think letting the math tell us what's there is putting the cart before the horse.

That's a strangely anti-science position from a rationalist. Generally we explain things in science by finding the relevant equations. Newton's contribution was not to say "Gosh, guys, I think things fall down"; it was to tie gravity to a specific, testable equation.

(You can do a lot of biology without math, of course. But the price paid is a loss of rigor. I'm a big fan of language, use it all day long, but it's full of pitfalls if you use only it when trying to explain things.)

In the case of QM, the state of this "matter" can only be got at through further experiments, neither from the math like Feynman nor metaphysical consistency like Cockshott.

You can't do better than the math in QM; it's almost all we have. It's so counter-intuitive that no non-math analogies can be trusted. You absolutely will go wrong reifying the interpretations— e.g. treating "wave form collapse" as a physical process.

If you want to try things to try to break the math— that's exactly what experimental physics is trying to do. Hopefully it breaks so hard that new theories are needed. (At this point we do need both new theories and new experiments— we know about dark matter but don't have any idea what it is.)

[rotting bones]

That was an appeal to empiricism, not an opposition to math. I'm the one who argued for more math education in schools focusing on apparently useless subjects like calculus, and I stand by that assessment.

That's a strangely anti-science position from a rationalist.

It's a pro-science position. If the math dictates the answer, why fund experiments?

Generally we explain things in science by finding the relevant equations. Newton's contribution was not to say "Gosh, guys, I think things fall down"; it was to tie gravity to a specific, testable equation.

Obviously, we should use math to write the theories. How would we keep track of the territory without legends on the map?

(You can do a lot of biology without math, of course. But the price paid is a loss of rigor. I'm a big fan of language, use it all day long, but it's full of pitfalls if you use only it when trying to explain things.)

Math is a language. It's just more precise than spoken language. Language is something that comes from humans, which is why it's not to be trusted. What is not a language is the experiment, and that's what I believe in.

You can't do better than the math in QM; it's almost all we have. It's so counter-intuitive that no non-math analogies can be trusted. You absolutely will go wrong reifying the interpretations— e.g. treating "wave form collapse" as a physical process.

QM is a physical process. The "wave form collapse" is a legend on the map. We should clearly distinguish between experimental results, which represent the state of matter, from the theories we invent to explain them. Otherwise, we might forget to prioritize the former over the latter.

If you want to try things to try to break the math— that's exactly what experimental physics is trying to do. Hopefully it breaks so hard that new theories are needed. (At this point we do need both new theories and new experiments— we know about dark matter but don't have any idea what it is.)

The current state of physics is known to be wrong because there are multiple inconsistent theories. The last time that happened, the foundations of physics were shaken. I don't think trusting the math of Newtonian physics would have done scientists any favors when they were trying to formulate more precise accounts of the state of matter like relativity or QM.

[zompist]

That was an appeal to empiricism, not an opposition to math. I'm the one who argued for more math education in schools focusing on apparently useless subjects like calculus, and I stand by that assessment.

That's a strangely anti-science position from a rationalist.

It's a pro-science position. If the math dictates the answer, why fund experiments?

Generally we explain things in science by finding the relevant equations. Newton's contribution was not to say "Gosh, guys, I think things fall down"; it was to tie gravity to a specific, testable equation.

Obviously, we should use math to write the theories. How would we keep track of the territory without legends on the map?

Well, this seems to be a motte-and-bailey situation... you say something nonsensical and provocative ("I don't believe in math") and then retreat to an actually defensible position ("we should use math to write the theories").

No one believes "the math dictates the answer". Physicists love experiments, that's why they spend billions of dollars on them.

The current state of physics is known to be wrong because there are multiple inconsistent theories. The last time that happened, the foundations of physics were shaken.

Did you forget the "experiments" part so soon? The late-19th-century consensus was broken because it didn't match new data.

[rotting bones]

Well, this seems to be a motte-and-bailey situation... you say something nonsensical and provocative ("I don't believe in math") and then retreat to an actually defensible position ("we should use math to write the theories").

I clarified immediately:

I believe that math is like the legends on a map that can be used to keep track of what's really there. "What's really there" is matter by definition. I think letting the math tell us what's there is putting the cart before the horse. In the case of QM, the state of this "matter" can only be got at through further experiments...

Here, I'm clearly prioritizing experimentation over math.

No one believes "the math dictates the answer". Physicists love experiments, that's why they spend billions of dollars on them.

No, but many believe an interpretation that's closer to the math is more likely to be correct. I don't believe that either for the reasons I explained.

Did you forget the "experiments" part so soon? The late-19th-century consensus was broken because it didn't match new data.

My understanding is that we normally don't have the resources to conduct experiments that bring our inconsistent theories into direct confrontation, but I could be wrong about that. What this tells me is that trusting the current state of the math is a fool's errand.

[Travis B.]

One thing to remember is that math is suspiciously accurate at forming predictions about the natural world for some reason. There have been many things where much of the math came before the observations that confirmed it, e.g. many things associated with general relativity. Even things where those who came up with the math's predictions were doubtful (e.g. gravitational waves) were later proven to be true from observations.

[rotting bones]

One thing to remember is that math is suspiciously accurate at forming predictions about the natural world for some reason. There have been many things where much of the math came before the observations that confirmed it, e.g. many things associated with general relativity. Even things where those who came up with the math's predictions were doubtful (e.g. gravitational waves) were later proven to be true from observations.

There is even more math that doesn't describe the physical world all that eerily. Those don't get pursued as much. E.g. Octonions.