Especially if what he’s saying has lacks a major hallmark of crackpot theories. That is he is very open to his theory being wrong and he won’t even say that it’s probably correct. He just thinks it’s possible and would like to see more work done on it.
The derogatory aspect of calling someone a crank is obviously uncalled for, but as a shorthand it’s not unreasonable to use the term.
Walter Alvarez is another Nobel laureate who proposed a theory that an asteroid wiped out the dinosaurs. Many people thought that was crank science.
And it’s important to note he had another out there idea that there were hidden chambers in the pyramids of Egypt. That one turned out to be wrong.
What distinguished his theories from crank science is that he was open to the idea that they were wrong and was interested in actually using the scientific method to investigate them.
The only thing that’s really not is the microtuble thing and he collaborated with someone else on that.
That’s also the least interesting thing he says I think because he admits it’s just an interesting place to look for quantum effects and he has no idea if he’s right.
Also, Penrose has not been "studying the topic since undergrad" ... he's been bothered by the notion that he is "just a computer" since then, but he didn't get into seriously addressing it until much later, and he's never studied it--he notoriously ignores the entirety of the literature of philosophy of mind and neuroscience.
> That’s also the least interesting thing he says I think because he admits it’s just an interesting place to look for quantum effects and he has no idea if he’s right.
This is simply not accurate.
And the fact is that Penrose is completely irrelevant to the subject of consciousness other than via an argument from invalid authority.
P.S. The response is disingenuous and discrediting ... I won't respond to that person again, especially after seeing this comment: "I’m in the Penrose camp that Turing machines can’t be conscious which is required for true AGI" --- this is pure ideology. TMs are clearly adequate for AGI even if somehow "TMs can't be conscious" ... c.f. Chalmers' philosophical zombies.
P.S. this response was merely a condensed version of he response to my reply. Here’s a longer one.
>Nothing that Penrose has talked about in re consciousness is "soundly within his field"
Quantum gravity, physics, and mathematical logic are clearly within his field. He’s a physicist, mathematician, and logician
>Also, Penrose has not been "studying the topic since undergrad" ..
He says that he has.
>he notoriously ignores the entirety of the literature of philosophy of mind and neuroscience.
Disagreement with some members of a field isn’t the same as ignoring it.
>This is simply not accurate.
It is. I heard it from the horses mouth.
>And the fact is that Penrose is completely irrelevant to the subject of consciousness other than via an argument from invalid authority.
It’s a field that has never produced anything concrete or beneficial to anyone outside the field. I’m unsure what relevance even means there.
>P.S. The response is disingenuous and discrediting ... I won't respond to that person again,
Yet you did. I saw your edit.
>Clearly TMs can be conscious
I guess that’s that then.
You just placed biology and neuroscience firmly outside of his wheelhouse. Just the size of the structures involved, temperature, timescales, and distance between neurons alone is a serious problem with his theory here.
If he is approaching things from a purely hypothetical standpoint it’s an unlikely but reasonable idea, but it utterly fails as part of how a larger system we actually understand quite a bit about works. Which is always the hard part of science, you’re not just fitting a single curve but thousands of different datasets.
For biology he had a collaborator. You aren’t likely to find many biologist / quantum physicists.
The term neuroscience wasn’t even coined until after he finished his PhD. You could probably name many other relevant sub specialties that doesn’t have formal training in.
If any of his theories are correct, you wouldn’t expect a neuroscientist, or biologist to be equipped to come up with them.
>If he is approaching things from a purely hypothetical standpoint it’s an unlikely but reasonable idea
He is. His first book essentially had no proposed mechanism. Then an anesthesiologist and researcher read it and contacted him with the proposal that microtubules might provide an environment that is insulated from the normal warm, wet, and noisy environment of the brain.
His next book investigated that idea, but he’s repeatedly said that this is just an interesting place to investigate and he has no idea whether it’s true.
>but it utterly fails as part of how a larger system we actually understand quite a bit about works.
How does it fail? I’ve read quite a hit about it and plenty for people are skeptical but I’ve never seen anyone showing how it “utterly fails”.
The first car wasn’t called a car by the people who built it, but we back date terms. He’s not a neuroscientist because he’s not studied the brain’s physical structures.
> insulated from the normal warm, wet, and noisy environment of the brain
? The tube is made of atoms at the same temperature as what’s outside the tube, there’s no isolation here.
> How does it fail?
It fails in many many ways. Individual neurons are vastly too small for consensus to occur on that scale you need something involving millions of them at body temperature at the very low end. Local quantum effects are obviously going on but they don’t scale.
Penrose has certainly studied the brains physical structures. He has 40 years of books and papers published on the subject.
>same temperature
No one is proposing that they are literally thermally insulating.
> on that scale you need something involving millions of them at body temperature at the very low end. Local quantum effects are obviously going on but they don’t scale.
That sounds like a good problem for a Nobel Prize winning physicist to investigate. If a neuroscientist were investigating it, I’d expect them to bring in a physicist.
> That sounds like a good problem for a Nobel Prize winning physicist to investigate.
No that’s a fairly trivial problem anyone with an understanding of QM can investigate. Atoms are atoms here it doesn’t really matter what biological structures are involved they are floating around in warm water.
If that’s the definition we’re going with, then anyone who does research that touches on neuroscience is a neuroscientist.
>trivial
Calling it trivial is hand-waving. Tegmark’s fast-decoherence bounds hinge on specific parameter choices; change the dielectric, charge model, spacing, or geometry and the timescales move into a regime that might matter. Temperature equality doesn’t erase structure. Ordered environments and collective modes can suppress decoherence without “insulating the brain.” Microtubules are a testable hypothesis, not a creed.
If you think they fail, point to a concrete model that rules out coherence under corrected parameters or shows a clash with measured neural dynamics and energy budgets. “Warm water, case closed” is an assertion, not that model.
It's a lie to claim that I said or even implied this.
OTOH, the "camp that Turing machines can't be conscious" is pure ideology and is based on repeatedly proven logic errors--Lucas was known to be wrong about Godel before Penrose came along and embraced his errors. And it's very common for people in that camp to project their own unsubstantiated baseless faith "that Turing machines can't be conscious" (which for Penrose, like many others in the camp, was a consequence of a semi-religious metaphysical notion that he wasn't "just a computer") onto rational informed people, with rhetoric like "people in that have unwavering faith in the idea that consciousness arises from computable processes" -- it's the logically default position, a consequence of intelligence and knowledge, not faith. The hilarious thing is that "consciousness arises from quantum effects" doesn't get the no-TM faithers what they want--they're still "just" machines, even if the machines use qubits rather than bits.
Penrose and Lucas’ argument may or may not be correct, but that still doesn’t imply that consciousness can arise from computable processes. There is no reason that it should. There is absolutely nothing to suggest this should be the default position.
The only way to get to this position is through faith. Doesn’t mean you’re wrong. But it’s not a falsifiable position since you can’t prove consciousness.
Keeping consciousness undefined means the requirements to form it are also undefined. There’s no way kizzip can arise from computers, there’s no way kizzip can arise from anything other than computers.
It is perfectly fine for you to adopt that belief. My issue is in declaring that belief to be self-evident support for calling someone a crank.
Why? There’s a lot of Nobel laureates over time who collectively made many such claims, so you can easily pick examples in both directions.
My point is more such ideas aren’t accurate enough for anything beyond preliminary testing by actual scientific investigation which sometimes does validate them but also commonly disproves them. There’s zero reason for the average person to consider their validity.
Your example is a perfect demonstration of why most people ignoring such things is a good idea, these things don’t simply disappear without investigation.
But you also shouldn’t go around immediately dismissing any theories they have as crank science.
Paying attention has a cost so what’s the payoff?
I just said you shouldn’t dismiss it as crank science.
Plenty of people think his theories are wrong or unlikely. The only people “dismissing them as crank science” are people in that have unwavering faith in the idea that consciousness arises from computable processes.