littlebabyjesus
one of Maxwell's demons
No it doesn't. A definition of 'universe' that would allow what you are saying would be like a definition of 'universe' that excluded everything we can't see with our binoculars.
Cosmological proof of God's existence
- Thread starter bhamgeezer
- Start date
No it doesn't. A definition of 'universe' that would allow what you are saying would be like a definition of 'universe' that excluded everything we can't see with our binoculars.
His argument does not rely on any particular incomplete formal system nor any particular undecidable statements within those formal systems, just that such formal systems exist.
Oh come on. Even some phycisists posit a multiverse of which ours is but one.
littlebabyjesus
one of Maxwell's demons
Yes. And properly speaking, such a conception is simply a universe.
Imagine a sentient being on a planet orbiting a star that doesn't yet exist in the Milky Way 100 billion years from now. If current cosmology is correct, such a being looking out into the sky will see only the Milky Way. All other galaxies will be so far away that the being will have no way to know they are there. Using best scientific method, they will conclude that the universe consists of what we now call the Milky Way surrounded by seemingly endless empty space. A clever scientist then comes along and uses the conception of other galaxies so far away that we can never know about them in order to explain some seemingly odd measurements.
That's what the conception of the multiverse is. It is an extension of what is commonly thought of as the universe to places that we cannot possibly have direct evidence for.
More or less, except that usually it is hypothesized that the different universes have different fundamental constants so that cosmologists can apply a version of the anthropic principle to explain the fine tuning of our universe.
Let's take a step back. With our scientific reductionist understanding of the universe we don't consider consciousness and intentionality to be fundamental features of the world. Rain doesn't fall in order to water the plants. We are used to bottom up explanations - plants have evolved to take advantage of the rain. If a phenomenon resists this sort of bottom up explanation then it means that either we have failed to find such an explanation but it could mean that our scientific reductionist ideas sometimes necessarily fail. Your conception of the universe is based on such a scientific reductionist scheme. If you reject that scheme you might also find a conception of the supernatural to be coherent and even meaningful. This realisation wouldn't rule out the scientific method as a useful method because you could still systematically test empirical hypotheses, you just wouldn't necessarily come to the type of naturalistic conclusion that the scientific method usually arrives at. Our concepts of nature and the universe are not prior to our investigation of nature and the universe.
littlebabyjesus
one of Maxwell's demons
Yes. My hunch, and it is only a hunch, is that we simply don't have a deep enough understanding to see that the cosmological constants have to be as they are in order for a universe to exist. That they cannot be different by definition – in the same way that π is the number it is by definition for all circles because otherwise they wouldn't be circles as defined.
But that's only a hunch, an indication that we shouldn't give up trying to find that understanding. Everything that can happen happens is an attractive short-cut, but it also sounds like a cop-out to me, certainly not an end to investigation. I might be wrong, though – my hunch is based on little more than an aesthetic preference.
Either way, my point still stands, though.
They necessarily 'fail' with regard to subjective experience, which is something we can only know to be true. We cannot test that knowledge, but we can be secure in it nonetheless. That's at the heart of everything I've been arguing so far.
That's not actually a failure, though. It simply is, and a full system will incorporate that fact. That's the challenge.
I don't reject your conception of the nature and supernature, I just reject what seems to be an a priori insistence that that's the way we have to conceive these matters. We have our way of thinking, our way of talking about these matters as a product of the success of this way of thinking not because other conceptions necessarily logically circular.
This little thread keeps triggering my mathematician's instincts. Most obviously with the use of Godel's incompleteness theorems. I don't mind informal or hand waving arguments, but there should still be an indication of how the formal proof should work. Then there's bhamgeezer's counterfactual causality with it's implicit infinite sum of all possible universes. Again the proof is in the pudding or rather the proof is in the proof. You need an indication of how you can rigorously perform such a feat - it's not obvious. Then there's this insistence of the emptiness of concepts such as "god". To solve a problem you need to look at things afresh. If you are stuck on a problem it's like your thought patterns can't escape themselves. Whenever you say something like, "obviously that can't be so", then you should count your blessings - you've discovered your mental block. For heaven's sake reject what you think is "obvious" until you've proved it!
Maybe non-mathematicians shouldn't be allowed to do philosophy. Not that mathematicians are any better, but that's a different story,
This little thread keeps triggering my mathematician's instincts. Most obviously with the use of Godel's incompleteness theorems. I don't mind informal or hand waving arguments, but there should still be an indication of how the formal proof should work. Then there's bhamgeezer's counterfactual causality with it's implicit infinite sum of all possible universes. Again the proof is in the pudding or rather the proof is in the proof. You need an indication of how you can rigorously perform such a feat - it's not obvious. Then there's this insistence of the emptiness of concepts such as "god". To solve a problem you need to look at things afresh. If you are stuck on a problem it's like your thought patterns can't escape themselves. Whenever you say something like, "obviously that can't be so", then you should count your blessings - you've discovered your mental block. For heaven's sake reject what you think is "obvious" until you've proved it!
Maybe non-mathematicians shouldn't be allowed to do philosophy. Not that mathematicians are any better, but that's a different story,
littlebabyjesus
one of Maxwell's demons
Perhaps a better line of attack, given that it would take me quite a bit of work to provide convincing proof that I'm right about Godel (I'm sure I am right, btw), would be to talk instead of the halting problem to any conceivable Turing machine.
I would contend that we subjective beings are able to solve the halting problem precisely because we are not Turing machines. How? We cannot know. We cannot have direct access to our non-algorithmic understanding and we cannot formalise it in any mathematical system as a means to model it, providing a model that we can look at. At least, we must abandon mathematics as our means of understanding. It is a different kind of understanding, one that cannot be expressed in language, be that words or mathematical language.
In other words, and I know you hate this terminology but I think it is accurate, we are unable to provide a complete model of ourselves that allows us complete understanding of ourselves. Moreover, we can prove that we are unable to provide a complete model. And it can only ever be this model that we can have direct access to.
ETA:
Also, I have to say that I find your last post to be a prime example of the very thing you are warning against – content-free handwaving.
I would contend that we subjective beings are able to solve the halting problem precisely because we are not Turing machines. How? We cannot know. We cannot have direct access to our non-algorithmic understanding and we cannot formalise it in any mathematical system as a means to model it, providing a model that we can look at. At least, we must abandon mathematics as our means of understanding. It is a different kind of understanding, one that cannot be expressed in language, be that words or mathematical language.
In other words, and I know you hate this terminology but I think it is accurate, we are unable to provide a complete model of ourselves that allows us complete understanding of ourselves. Moreover, we can prove that we are unable to provide a complete model. And it can only ever be this model that we can have direct access to.
ETA:
Also, I have to say that I find your last post to be a prime example of the very thing you are warning against – content-free handwaving.
I don't hate what you've said just now. I think you should make a distinction between following an algorithm and understanding something (there's a bit of a category error there). What you've just said is a hand waving sketch of a proof that might just work. What you said before seemed hopelessly wrong though.
I should say that if you agree with Penrose then their shouldn't be a Godel statement about the functioning of our brains that we cannot understand fully because our brains are not Turing machines according to Penrose!
I should say that if you agree with Penrose then their shouldn't be a Godel statement about the functioning of our brains that we cannot understand fully because our brains are not Turing machines according to Penrose!
littlebabyjesus
one of Maxwell's demons
I agree with Penrose that we are not (indeed cannot be) Turing machines. I actually agree with Penrose about a lot more than I did when I first read him!
However, I'm not so sure your last statement is correct, but I may be misunderstanding the parallels between the proof that no Turing machine can solve the halting problem and Godel's theorem.
I will go away and research this a little as I am still sure I'm right about this. I think the key is the 'understand fully' concept. By 'understand fully', I mean explain fully. I have already said that we can know it to be true. However explanations fall into the category of 'model'. It is at this stage that it is a mistake to include any 'god'-style concept in the explanation. Ultimately we cannot explain how we know. We cannot explain how we solve the halting problem. We cannot explain how we are not Turing machines. There are limits to the extent to which we can model ourselves.
However, I'm not so sure your last statement is correct, but I may be misunderstanding the parallels between the proof that no Turing machine can solve the halting problem and Godel's theorem.
I will go away and research this a little as I am still sure I'm right about this. I think the key is the 'understand fully' concept. By 'understand fully', I mean explain fully. I have already said that we can know it to be true. However explanations fall into the category of 'model'. It is at this stage that it is a mistake to include any 'god'-style concept in the explanation. Ultimately we cannot explain how we know. We cannot explain how we solve the halting problem. We cannot explain how we are not Turing machines. There are limits to the extent to which we can model ourselves.
I'm reluctant to recommend it because I think Penrose is wrong on this, but you should try reading Shadows of the Mind. As it happens one of the alternatives that Penrose offers is a "god" style solution - the brain is a sort of ineffable Turing Machine. You might also try reading Beyond the Doubting of a Shadow, which is online somewhere. This latter is a response to critics and includes links to their critiques, most of which are worth reading as well.
I think Penrose's argument relies on switching between consciously knowing a statement is true in some sort of objective sense on the one hand and being able to follow an algorithm which produces a formal proof of the statement on the other.
Do you see 2+3=5 as being true? If somebody denied it, they wouldn't be so much as wrong but rather playing the wrong sort of game in arithmetic. I don't think that I am consciously aware that 2+3=5, such facts are a product of how I learnt how to do arithmetic. To not know it would be like not knowing how to ride a bike rather than not knowing what the capital of France is.
I think Penrose's argument relies on switching between consciously knowing a statement is true in some sort of objective sense on the one hand and being able to follow an algorithm which produces a formal proof of the statement on the other.
Do you see 2+3=5 as being true? If somebody denied it, they wouldn't be so much as wrong but rather playing the wrong sort of game in arithmetic. I don't think that I am consciously aware that 2+3=5, such facts are a product of how I learnt how to do arithmetic. To not know it would be like not knowing how to ride a bike rather than not knowing what the capital of France is.
littlebabyjesus
one of Maxwell's demons
Ah, this is the problem. It's a bit like when philosophers start talking about 'free will'. They quickly descend into utter nonsense. So it is here, I think. It may be easy to show that we are not Turing machines, but it's far harder to even suggest what we might be instead.
I'll try your recommendations, though.
What do you think Penrose is wrong about, that we are not Turing machines, or the suggestions about what we might be instead?
I'll try your recommendations, though.
What do you think Penrose is wrong about, that we are not Turing machines, or the suggestions about what we might be instead?
It's definitely not easy to show that we are not Turing machines! It's like proving a negative. You've got to show that no Turing machine can perform a certain task. It's why Penrose's argument is so involved. That's the other reason I'm reluctant to recommend it. It's such a headache!
littlebabyjesus said:I'll try your recommendations, though.
What do you think Penrose is wrong about, that we are not Turing machines, or the suggestions about what we might be instead?
I think his reasoning about why we can't be Turing machines is flawed ie. his Godelian argument. The physics is fine as far as I can tell, but it doesn't make for a plausible case - quite the reverse! Even if there are quantum mechanical processes that have a central role in the functioning of the brain, you still need to somehow exploit a hypothetical quantum gravity computational resource which nobody has any working theory for yet.
littlebabyjesus
one of Maxwell's demons
You have to know when to stop.
It seems likely to me that we cannot prove that we are not Turing machines.
However, we can know that we are not Turing machines. And we can show it, can we not, simply by doing things that Turing machines cannot do. That's the right approach, I would think – you show that a Turing machine could not solve a particular problem, then you show that we can solve it. Isn't that the heart of Penrose's example of non-algorithmic understanding of Goodstein's theorem?
camouflage
gaslit at scale.
I'll stop you there, philosophers always talk utter nonesense.
monkeybrain hurls itself pitiably against the un-monkeybrainable, (and come to monkeybrain it... the un-monkeyspeakable). thus, nonesense.
same thing religion imo.
I really struggled with New Mind. Penrose is very smart and a lovely man to chat with about almost anything else.
But here, it's as if his thought process went:
- What's a mystery?
- Consciousness is!
- How to explain it?
- Quantum stuff - that's a mystery too!
That, and taking rather too seriously the nanotube enthusiast from Tucson whose name escapes me for the moment... Hameroff, that's it.
littlebabyjesus
one of Maxwell's demons
I agree, laptop. TBH, I love that book and have read it several times – all but the last part where he attempts to provide his quantum explanation. It isn't really an explanation at all.
However, having said that, I do think there are many aspects of our brain function that do require quantum explanations. For instance, the 'binding problem' I believe will be solved by showing that the brain takes an instantaneous 'snapshot' of itself through entanglement (and this is why neurons for consciousness fire simultaneously) in much the same way that algae use entanglement to transfer energy efficiently. The 'energy-efficient' path through the brain as 'truth'.
It would be surprising indeed if all processes requiring speed or efficiency did not involve such quantum effects. I think we will find that they are ubiquitous. But it is precisely because we do not 'model' it in our representation of reality that we have not found it before.
However, having said that, I do think there are many aspects of our brain function that do require quantum explanations. For instance, the 'binding problem' I believe will be solved by showing that the brain takes an instantaneous 'snapshot' of itself through entanglement (and this is why neurons for consciousness fire simultaneously) in much the same way that algae use entanglement to transfer energy efficiently. The 'energy-efficient' path through the brain as 'truth'.
It would be surprising indeed if all processes requiring speed or efficiency did not involve such quantum effects. I think we will find that they are ubiquitous. But it is precisely because we do not 'model' it in our representation of reality that we have not found it before.
littlebabyjesus
one of Maxwell's demons
I'll stop you there, philosophers always talk utter nonesense.
monkeybrain hurls itself pitiably against the un-monkeybrainable, (and come to monkeybrain it... the un-monkeyspeakable). thus, nonesense.
same thing religion imo.
Well, yes.
I don't.
What is an 'explanation of existence'?
What is the Godel statement of mathematics?
littlebabyjesus
one of Maxwell's demons
It is hard to demonstrate that mathematics is the best, most truthful way to describe the universe. It is, though. It is the best way to describe relationships between things, which is the only sensible approach to describing the universe, to describing cause and effect. As was said by Crispy on another thread, best not to ask what matter is, better to ask what matter does. This applies to everything – we cannot say what the universe is, merely what it does.
This article by Hawking begins to describe how Godel's theorem can be applied to other systems, and boils Godel down into its essence, that of self-reference:
Penrose examines this in far greater detail, and also considers the similarity between this reasoning and Turing's proof that no Turing machine could ever solve the stopping problem. It has similar reasoning. For instance, consider the Goldbach conjecture that every even number greater than 2 is the sum of two prime numbers. You set a machine to try out each even number in turn and to stop when it comes across one that is not the sum of two primes. But what if the conjecture is true? You then need another algorithm that will decide whether or not this algorithm will ever stop. And such a 'stopping algorithm' does not exist. (You can prove that it doesn't exist, but I'm not going to go into that; you'll have to take my word for it for the sake of brevity.) You need something from outside the set of algorithms to decide upon the validity of the algorithms – sets of algorithms cannot justify themselves on their own.
Sorry if this is an incomplete explanation, but I hope it is pointing in the direction of my thought. TBH if you don't accept my initial statement about mathematics, none of the rest follows at all.
This article by Hawking begins to describe how Godel's theorem can be applied to other systems, and boils Godel down into its essence, that of self-reference:
Penrose examines this in far greater detail, and also considers the similarity between this reasoning and Turing's proof that no Turing machine could ever solve the stopping problem. It has similar reasoning. For instance, consider the Goldbach conjecture that every even number greater than 2 is the sum of two prime numbers. You set a machine to try out each even number in turn and to stop when it comes across one that is not the sum of two primes. But what if the conjecture is true? You then need another algorithm that will decide whether or not this algorithm will ever stop. And such a 'stopping algorithm' does not exist. (You can prove that it doesn't exist, but I'm not going to go into that; you'll have to take my word for it for the sake of brevity.) You need something from outside the set of algorithms to decide upon the validity of the algorithms – sets of algorithms cannot justify themselves on their own.
Sorry if this is an incomplete explanation, but I hope it is pointing in the direction of my thought. TBH if you don't accept my initial statement about mathematics, none of the rest follows at all.
The thing is, even if maths is the 'best and most truthful' way of describing the universe, that does not even begin to entail that the universe is subject to some Godelian incompleteness. You need to go much further and posit a kind of Platonic relationship between maths and reality, so that maths is actually the underlying code of the universe. But that just brings you back to the same old Platonic error.
littlebabyjesus
one of Maxwell's demons
No. Because I've stated throughout that all we have access to is our model of the universe, which we check against measurements. Our explanations of the universe are subject to incompleteness. And our explanations are all we have access to.
Most physicists accept this, that there is no such thing, ultimately, as the truth. All there are are good theories and better theories. That's the process of science.
Feynman described this rather beautifully as an enquiry into chess where you don't have a rulebook, and you are trying to work out the rules by studying games of chess. First you note that a bishop can move but cannot change its colour. Then you discover a deeper truth, which is that a bishop can only move diagonally. This seems to work for a long time, then one day, you observe a bishop appearing on the opposite colour to the one left on the board, and discover that a pawn can become a bishop in certain circumstances. Each game provides more evidence, but you never reach a time when you can conclusively state that you have categorically solved the game of chess.
Most physicists accept this, that there is no such thing, ultimately, as the truth. All there are are good theories and better theories. That's the process of science.
Feynman described this rather beautifully as an enquiry into chess where you don't have a rulebook, and you are trying to work out the rules by studying games of chess. First you note that a bishop can move but cannot change its colour. Then you discover a deeper truth, which is that a bishop can only move diagonally. This seems to work for a long time, then one day, you observe a bishop appearing on the opposite colour to the one left on the board, and discover that a pawn can become a bishop in certain circumstances. Each game provides more evidence, but you never reach a time when you can conclusively state that you have categorically solved the game of chess.
That's either a bad analogy, because chess is a finite game so that eventually you would have a description of every possible chess-event, or it's a good analogy because chess is a finite game so that eventually you would have a description of every possible chess-event.
I've got no problem with there being no such thing as The Truth, for a given definition of Truth of course.
Can you explain what you mean by having 'access' to the universe/our model of the universe. Because that also suggests a perennial error.
I've got no problem with there being no such thing as The Truth, for a given definition of Truth of course.
Can you explain what you mean by having 'access' to the universe/our model of the universe. Because that also suggests a perennial error.
littlebabyjesus
one of Maxwell's demons
All possible content of consciousness is a result of neural states in one form or another, not a result of anything 'outside'. We literally model the universe, building up layer after layer of meanings, such as colours, shapes, notes, size, movement, everything. We decide which of these to include in our consciousness according to various factors, primarily of course that which is picked up by our senses, but we're perfectly capable of running a full model with no external input at all.
Ultimately this is all we have access to, and for a long time it was the only thing we made theories about. It's only fairly recently that we've even started to consider things that we cannot model in consciousness and to think about them and conceptualise, for instance, waves we cannot see or spacetime we cannot unify in our consciousnesses. In the end we can only understand higher dimensions in the language of mathematics. Our brains are not capable of modelling them – we have evolved to be able to move around in the three dimensions of space that are appropriate to our scale, after all, and with time and space as separate categories, which is appropriate to the speeds at which we travel.
It is to the huge credit of humans, I think, that we've been able to start to glimpse what lies beyond our immediate perception, but we can only do this through abstraction. Thus we have moved from examination of what we perceive to consideration of what we cannot and to an abstract conceptualisation of it. But we're left in the same essential position. We can only consider our abstractions about, for instance, the electron. We cannot consider electrons directly. We can test how well our abstractions fit through collecting data, but that data is only ever more than noise once we have applied our meanings to it, whether through neural processing to produce conscious images, or through the processing of our abstractions. It is the results of that processing, and only those results, that we can look at.
Ultimately this is all we have access to, and for a long time it was the only thing we made theories about. It's only fairly recently that we've even started to consider things that we cannot model in consciousness and to think about them and conceptualise, for instance, waves we cannot see or spacetime we cannot unify in our consciousnesses. In the end we can only understand higher dimensions in the language of mathematics. Our brains are not capable of modelling them – we have evolved to be able to move around in the three dimensions of space that are appropriate to our scale, after all, and with time and space as separate categories, which is appropriate to the speeds at which we travel.
It is to the huge credit of humans, I think, that we've been able to start to glimpse what lies beyond our immediate perception, but we can only do this through abstraction. Thus we have moved from examination of what we perceive to consideration of what we cannot and to an abstract conceptualisation of it. But we're left in the same essential position. We can only consider our abstractions about, for instance, the electron. We cannot consider electrons directly. We can test how well our abstractions fit through collecting data, but that data is only ever more than noise once we have applied our meanings to it, whether through neural processing to produce conscious images, or through the processing of our abstractions. It is the results of that processing, and only those results, that we can look at.
littlebabyjesus
one of Maxwell's demons
You may have a description of every possible chess event after watching enough games of chess, but how would you know that you had a description of every possible chess event? You could hypothesise that you had a full description. You may or may not be able to prove that hypothesis. I'm not sure, tbh. It is not immediately obvious to me that you could.
For instance, to take my bishop example, generally speaking there is no reason ever to promote a pawn to a bishop. A queen is always a better option.* You can, though, and one day a game could arise in which it makes no difference – it's checkmate next move either way, so there is no way to decide which piece to choose. There is more to chess than its rules – there is discretion for the player. How do you account for that? You might conclude from watching so many games that each player makes the optimum move based on x amounts of computations. But one player might be a human. Humans use their aesthetic sense to decide what move to make. Through experience they come to know that certain shapes are good while others are bad. There are all kinds of potential barriers to your knowing, or even suspecting, that you've 'solved' the rules of chess.
*Almost always. It is conceivable that you would want not to put your opponent's king in check that move because you want your opponent to leave their king where it is, or that, more likely, promoting to a queen would produce a stalemate. It's quite a far-fetched situation, though – although that's a good illustration of how hard it is to know that you've come to the end. And you'll probably have to watch a lot of games for this situation, one in which you promote to a rook and one in which you promote to a knight all come up. Then you're left with the problem of working out whether or not you are in fact allowed to promote to a queen if that would produce a stalemate. How do you answer that one? Do the rules allow you to move your king into check, or is that simply something you've never seen because it's never the right move? There are all kinds of problems and barriers to complete knowledge.
The Goldbach conjecture is an open problem, there is no reason to think that it is inherently undecidable. You're missing the argument. Maybe look at Cantor's diagonalisation argument, which is similar but simpler.
Cantor proved that the set of real numbers (all numbers with decimal expansions including whole numbers, fractions, irrational numbers, transcendental numbers such as Pi and e) cannot be put in a one-to-one correspondence with the natural numbers (positive whole numbers).
His proof is a reductio ad absurdum. Suppose we could list every real number so there is a real number in the first place, another in the next and so on until we have a complete list including every real number. We then construct a new real by taking the first decimal digit from the first member of the list, the second decimal digit from the second number in the list, the third decimal digit from the third number on the list and so on until we have an infinite decimal expansion. We construct a new decimal expansion by taking changing each digit in this expansion by replacing it with a 0 if it was not already a 0 or otherwise a 1 if it was a 0. This decimal expansion cannot be on the list and so the set of reals is not "countable".
The proof of Godel's theorem uses a similar diagonal construction. Note that Cantors construction is very artificial and entirely dependent on the arbitrary nature of the how we choose to try to list the real numbers. Godel's construction is similarly artificial and dependent on the the formal system we use to try to capture all proofs in arithmetic. Godelian arguments have a certain self-referential character - like the liar paradox. They aren't just open questions. In fact the very construction of the Godelian statement is all but a proof of the statement even if the formal system itself cannot contain the proof.
Here's Penrose expressing his reasoning succinctly:
http://www.calculemus.org/MathUniversalis/NS/10/01penrose.html
I think it is quite clear where Penrose's mistake lies. When we "see the truth" of a mathematical statement it is not usually because we have followed the exact details of the proof. When you try to break down mathematics to it's most basic axiomatic truths from which you base all derive all other truths then you find that the axioms are quite odd and even in some cases disputed. The axiomatic formality of mathematics does not correspond to our basic understanding, which is learnt in primary school. You don't teach 5 year olds the Zermelo Fraenkel axioms in order to teach them that 2+3=5. What we recognise as true is what we recognise by a convincing demonstration. But there needn't be an absolute gold standard of what a convincing demonstration is. Our brains could be Turing machines but not Turing machines that recognise the truth of statements because they are in accord with a formal system. There needn't be a one-to-one correspondence between brain state and mental state. Penrose's problem is that he is a hard AI man with a twist that hard AI isn't powerful enough.
littlebabyjesus
one of Maxwell's demons
Yes. I've been kind of trying to say this, but it seems failing. I realised I'd made a mistake in not seeing this in another thread, but have been trying to correct it in this. It's why we don't need god, basically.
Ok, but that means that our brains are Turing machines +, with added heuristics, if you like. Or added emotion.
That's an interesting last point about Penrose. I'll need to think about that.
His argument is a Sherlock Holmes argument. If you can eliminate what's impossible then what remains is the case regardless of how improbable it is. Sounds like genius? Remember Arthur Conan Doyle believed in fairies.
To be fair, Penrose doesn't just pick on quantum theory because it's mysterioius. He eliminates all the other physical theories because they are either computational or implausible in their application to the functioning of the brain (general relativity). He also eliminates quantum mechanics btw. His hypothesis is quantum gravitational and he doesn't even have a theory for how it works.
Similar threads
