Google engineer put on leave after saying AI chatbot has become sentient

Pages: 123
you have definitions of these terms that you feel very strong about.
Definitions are definitions. They aren't the conclusion of a line of reasoning, they're the starting point. You can disagree with a definition, like with any other axiom, but that just means we can't discuss the topic in question, because we can't agree on common terms to communicate with.

Particularly that if one excepts free will they must except soul.
That's not a definition, that's a conclusion. And no, I don't "feel very strongly" about it. It's a conclusion I've reached after careful consideration, and I'm more than willing to overturn it as soon as I find a compelling reason to do so.

I'm not sure for all events in the physical world cause is understood.
It's not necessary to answer my question. The question wasn't about the particular causes of given event, but about whether there are any other events besides thoughts that don't have physical causes.

Incidentally, it's epistemologically impossible to determine with certainty the causal relationship between two events. If A happened, then B happened, it might be that A caused B, or it might be that an unknown C actually caused B in such a way that it appears that A caused it, and you would have no way of knowing.

The Big-bang explains a lot, but I'm nearly certain it's cause is unknown.
If the Big Bang is the very start of space-time then it's not that it's divorced from causality, but rather it's the start of all causal chains. My question was whether there's any other process smack down in the middle of causality that is uncaused.
The clock might also think "it's my decision to do this. It just so happens to align exactly with the physics of my mechanism." It seems to erode the term into meaninglessness.

It's not my decision to like ice cream. However, given that being a preference, my choice to consume ice cream at some point would be my own decision.

This decision is not unpredictable.

I mean, it's the definition of free will

As I've read it, "Free will is the capacity of agents to choose between different possible courses of action unimpeded." Nowhere does it say free will must be free of PHYSICAL PROCESS. That would require not having a brain.

That's special pleading. I asked what makes thoughts special and the answer is that thoughts are special because they're thoughts?

Special is not objective, so I'm not sure what answer you wanted. Obviously thoughts are pretty special since more than 99% of the universe, all matter, and everything ever doesn't do it.

The decisions that the universe dictates you will make are not "against your will" because your will is also dictated by those same laws

But to say our will came from physics is a poor explanation when we know deeper laws that have caused evolution which generated our brains. Our brains run on physics, sure, but the physics is running a poorly coded program in our brain.

The real question is, is there a program that you could write that would decisively create a sentient AI with "free will"? If the logic of creating free will is possible, then it doesn't matter if it runs on physics.


If your capacity to decide is not merely conditioned, but entirely determined by the laws of physics, then you have no free will.

It does not mean that free will cannot come from physics, just that through physics you can theoretically determine the outcome of free will. With enough capacity for information and computation, you could explain anything through purely physical means, but you wouldn't get an explanation that actually makes sense in context.

"He ran away from the tiger because... particles in a certain region of the brain vibrated at a certain rate?"

Through evolution, the brain developed in order to increase survival odds. The brain is a tool in which decisions can be formed. In order for the decisions to align with survival, survival instincts came first, then decision forming mechanics.

Psychology is not a study of physics, but the mind. We know we can condition, teach, and influence people's minds. Physics gives rise to the mechanics which enable consciousness and thought.

Our brain surely runs on physics just as everything else, but physics builds things bigger than itself, as we've always seen. You don't explain biology through physics. You can't use the explanation of quantum mechanics in order to explain classical physics, even though we know physics must also run on both and classical physics explains objects that run on quantum mechanics at the lowest level.


This was a nice read:

https://physicsworld.com/a/why-free-will-is-beyond-physics/


"The ability to reduce everything to simple fundamental laws does not imply the ability to start from those laws and reconstruct the universe.. The behaviour of large and complex aggregates of elementary particles, it turns out, is not to be understood in terms of a simple extrapolation of the properties of a few particles. Instead, at each level of complexity entirely new properties appear, and the understanding of the new behaviours requires research which I think is as fundamental in its nature as any other."



Religious people will read this laughing.
"Free will is the capacity of agents to choose between different possible courses of action unimpeded." Nowhere does it say free will must be free of PHYSICAL PROCESS.
It's implied in the definition, for the reason I explained immediately after the passage you quoted.
Google "free will is incompatible with determinism". I'm not bringing a new idea, reconciliating the intuitive sense that we are in control of our actions and that the universe is ruled by determinism is an age-old question. The current consensus is that free will is an illusion.

Special is not objective, so I'm not sure what answer you wanted.
Special is objective. The Sun and Jupiter are pretty similar in terms of chemical composition, but there's something in the Sun that makes it special, compared to gas giants like Jupiter. What makes thoughts special?

Obviously thoughts are pretty special since more than 99% of the universe, all matter, and everything ever doesn't do it.
That's an unfalsifiable statement. Yes, a rock very probably has no thoughts (or at the very least, the mechanism that would support those thoughts is very well hidden), but there's no possible way to know if, say, an electronic computer has no thoughts, but just no way to express them.

But to say our will came from physics is a poor explanation when we know deeper laws that have caused evolution which generated our brains.
I have no clue what this means.

Our brains run on physics, sure, but the physics is running a poorly coded program in our brain.
The quality of the coding is irrelevant. If you reach an incorrect conclusion, you will reach it according to the laws of physics that govern the particles in your brain. If you reach a correct conclusion you will reach it accoding to those same laws.

The real question is, is there a program that you could write that would decisively create a sentient AI with "free will"? If the logic of creating free will is possible, then it doesn't matter if it runs on physics.
You've made it clear that you call "free will" something different than I do, and I've already said free will is impossible, so I'll need you to clarify this question.
But, I will say that I think it's theoretically possible to completely emulate a human brain electronically.

You don't explain biology through physics. You can't use the explanation of quantum mechanics in order to explain classical physics, even though we know physics must also run on both and classical physics explains objects that run on quantum mechanics at the lowest level.
You're confusing the terms. When I say "physics determines your actions" I don't mean the science, I mean the lowest level interactions that govern the universe. The limit of our knowledge does not allow us to study the motion of planets, or the migration of european swallows, through quantum mechanics, but that does not mean that those processes are not causally linked. Ultimately, all events in the universe must be reducible to interactions between particles. It cannot be any other way, because all that exists is made of those particles. There's no level of magnification where a hypothetical Theory of Everything ceases to apply and biology starts to apply, but there are levels of magnification where the quantity of interactions between subatomic particles is so large that it would simply be impractical to continue the study at that level of detail. In other words, chemistry, biology, psychology, sociology, etc. exist not because physics lacks explanatory power for processes in question, but to keep the level of detail manageable to humans.

To put in terms a programmer can easily understand: suppose you're developing an open world RPG, and you find that sometimes a quest bugs out. Would you like to debug the program by hooking up an oscilloscope to the CPU, or would you prefer to do it at a higher level of abstraction? Yet, it's undeniable that the program's behavior is entirely determined by the behavior of the electronics, right?

This was a nice read:

https://physicsworld.com/a/why-free-will-is-beyond-physics/
This honestly reads like an elaborate strawman. Are there seriously people out there saying that since free will doesn't exist we don't need neurology? If there are, that seems like such an inane contention that it doesn't need refuting.
It's implied in the definition

This is the very thing we are debating, whether being free of influence would necessitate being free from the material world.

Special is objective

Special is only objective in a very specific context. Yours definitely wasn't specific enough.

That's an unfalsifiable statement

So we're not going to assume objects don't have thoughts until otherwise proven? I'm on the fence about Zeus myself.

There's no level of magnification where a hypothetical Theory of Everything ceases to apply and biology starts to apply

A Theory of Everything may or may not exist. The idea that everything is reducible to physics is unproven.

The thing to consider, of course, is whether particles that obey physics can come together to create something which must also abide by new laws. If this happens, those new laws will not allow the object to be reducible to "just" physics.

I believe the general consensus on that for now is "no", you probably can't reduce biology and every other field of science to physics. It could be a lack of knowledge in the field, or it could be as simple as things are not merely the sum of their parts sometimes.


Or, to put it in terms a programmer can easily understand: Suppose every particle is a class. When they come together they form a new class. This new class inherits from all the particles it's made up of. This new class is not reducible to any of its parts, you can only view this class as a whole.


It would be interesting if you could actually, through looking at particles alone, be able to tell someone's thoughts, feelings, and memories. If you could look at a collection of particles and know exactly what information they contained simply by their organization and vibrations.


Are there seriously people out there saying that since free will doesn't exist we don't need neurology?

No. I skimmed over the article. The part I quoted was the part I liked most. It did say some questionable ideas earlier, but clarified itself later.



Ouch. He closed his account :(
Last edited on
[redacted]

I've said my piece, Pax out to ya, peeps :)
Last edited on
Pfft. What a wimp. If you can't take the heat, get out of the kitchen.


The thing to consider, of course, is whether particles that obey physics can come together to create something which must also abide by new laws. If this happens, those new laws will not allow the object to be reducible to "just" physics.
Okay, let's consider this. When quarks join together they form neutrons and protons. Neutrons and protons are bound by physics, same as quarks. When protons and neutrons join, they form nucleii, still bound by physics. Nucleii joined with electrons form atoms, also bound by physics. Bonded atoms form molecules, still bound by physics. Molecules can be organized to varying degrees of structure to form various materials (e.g. carbon organized one way can be graphite, or diamond in another). Materials are still bound by physics. Materials interacting with any of the previously mentioned particles encompasses the totality of human experience, from electrons flowing through a copper wire to material stresses breaking your bones, as well as the electrochemistry in your brain that's processing my words. What's left out?
That we know of, there doesn't appear to be any hidden complexity at the higher organization levels that should lead us to believe that brains are fundamentally different from, say, digital clocks. When I say "fundamentally different" I mean that they respond to incomparable sets of physical laws. If there's going to be a revolution, it's going to be in our understanding of the subatomic world.

I believe the general consensus on that for now is "no", you probably can't reduce biology and every other field of science to physics. It could be a lack of knowledge in the field, or it could be as simple as things are not merely the sum of their parts sometimes.
I don't think so. But I do think that anti-reductionism is probably more prevalent in the softer sciences.

Suppose every particle is a class. When they come together they form a new class. This new class inherits from all the particles it's made up of. This new class is not reducible to any of its parts, you can only view this class as a whole.
Are we talking about inheritance or composition? I'm not aware of any is-a relationships in physics. A proton is not a quark, and a molecule is not an atom.

It would be interesting if you could actually, through looking at particles alone, be able to tell someone's thoughts, feelings, and memories. If you could look at a collection of particles and know exactly what information they contained simply by their organization and vibrations.
You could. You'd "just" need to simulate the universe to reconstruct that information, and then have sufficient understanding of the brain to know which bits of data to ignore and which to look at, and in which order. It's like how having a memory dump of a computer doesn't tell you anything if you don't know the CPU that was executing it.
Pfft. What a wimp. If you can't take the heat, get out of the kitchen.

BUT IM MAKING SPEGHATTI

Neutrons and protons are bound by physics, same as quarks... still bound by physics..

Just as if you were inheriting from a class. The components you inherit and final outcome is still bound by physics. However, that doesn't mean they're reducible to physics.

If there's going to be a revolution, it's going to be in our understanding of the subatomic world

Which we already know brings a lot more questions than answers.

I stumbled upon Hempel's dilemma while doing some research. Basically, it states that, when claiming everything can be reduced to physics, that ]must mean they think either everything is reducible to the physics that we have now (which is incomplete) or some "future" complete physics (which is unknown and would cause circular reasoning -> things are reducible to physics because future physics says so).

I also found this very nice read on how evolution itself (biology) may be a contradiction to the idea of everything reduces to physics:

https://www.talkorigins.org/faqs/evolphil/reduction.html

It provided what I think is a great example of how something complex, while definitely built upon physics, cannot be simply reduced to physics. Again, going with the quote from earlier that just because everything runs on physical laws does not mean you can start with physical laws and reconstruct the universe.

While there are many reductionalists I'm sure, I don't think it's fair to say that current science would agree with this view. At the very least, this view should not be a given.


It's also important to note that any new discovery on how things may interact at a higher level would not somehow go against physics. Rather, it will be brought into physics as another aspect.

Either way, we're debating something that is apparently very hotly debated in the scientific community.


Odd idea... popped into my mind. We sometimes like to assume something like X is true if we had infinite computing power and memory available. If something truly takes infinite amounts of computing power to be true, that would practically, and perhaps then literally and theoretically, be impossible to be true.

Something may seem to work out exceedingly well in theory, but the very fact that it requires infinite resources is like needing to divide by 0 in order to make it true, it simply means that the basis of the theory is flawed.

The idea that you can use physics to properly see the future would require lots of computing power and memory. Since the universe is infinite, it would also require infinite computing power and memory.

Not a flawless idea, but one I've been rolling around in my head.
Last edited on
Well, jeez. Now I'm thinking about all this crazy nonsense.

Back in high school I watched an anime called Gantz. The anime had this machine that could take your body, break it up into atoms, then reconstruct you somewhere else.

This got me thinking about how if you made a perfect copy of someone, you made a new person. Our conscious minds would not interlink and connect. An exact copy right down to the atom would still be a different person consciousness-wise.

This is the very reason why the idea of escaping death by "leaving" your brain will never work. You are your brain.

If the atoms that construct your brain were exchanged with new atoms, you've essentially died and been replaced; because you can take the old atoms and reconstruct them how they were, and you'd get the same person but with a different consciousness.

The science doesn't seem to be too clear on this, or perhaps I just can't find the info, but atoms in our brains do get replaced at least a little bit. The interesting question is, "does this matter?" Is there a certain amount of atoms that don't matter? Are there certain atoms that don't matter?

Have you already died? The brain cells themselves don't get replaced, but there is blood flow with atoms going in and out. Perhaps these atoms aren't essential for our consciousness, or maybe they are too little to make a significant dent enough to say we've really "died" before.



Anyway. There's no real way to know if you consciousness can be reduced to just physics without emergent properties that can only come as seeing it as a whole - and not the sum of its parts. This is because quantum mechanics is still being researched and how consciousness works is still a mystery.

However, it seems very improbable to me that by only considering physics and atoms that you could explain consciousness.

There's three bodies of thought so far:

Physicalism: Implying everything is reduced to physics and conscious is not real.

Or that everything is physical and that can give rise to consciousness, but consciousness cannot affect the outcome of anything therefore useless

Or that consciousness can have an effect as an emergent property. Not just a property that becomes obvious when physical things come together, but only possible when they do.


Oh well.
Last edited on
I don't yet have time to respond to the rest, but this is quick enough:

Back in high school I watched an anime called Gantz.
The manga is much better. It really hit its stride just after the anime ended. The ending is lame, though.

By the way, Gantz is inconsistent with its treatment of pseudo-teleportation. It's explicitly stated that the process is Star Trek-like. I.e. it deconstructs your body and reconstructs it someplace else. Such a process would kill you every time. However, characters are seen sending information back and forth between the two locations while the process is ongoing, whereas in reality at the speed it goes it would likely be a horrifying experience.
Oh, well, sufficiently advanced technology etc.

This is the very reason why the idea of escaping death by "leaving" your brain will never work. You are your brain.
Nah, it can be done, in principle. If you gradually replace your brain with electronics using nanomachines, your consciousness would move onto dryware without killing you. Then it's just a matter of transferring your thought processes someplace else without stopping them, basically by reassigning neurons onto a remote machine one by one. During the transfer process your consciousness would be spread over a large area and some parts of your brain would have a higher latency than the rest. Subjectively it would probably feel like a weird trip. At the very end of the process your original cyberbrain would be running at 0% utilization, save for linking your remote consciousness to your body. At that point you'd just need to be connected to a new set of senses.
The manga is much better. It really hit its stride just after the anime ended. The ending is lame, though.

That's what I hear. The anime ended abruptly cause it was getting cancelled which really sucked. I don't really read manga though, prefer having voice actors, music, and animation!

If you gradually replace your brain with electronics using nanomachines

You mean transform the atoms that make your brain from organic matter to electronics?! Never thought of that, could be impossible to do without adding extra atoms in. That might... delude (?) your consciousness, create something new.

Even if done with only your brains atoms, it would have to perfectly simulate the working conditions of the brain or your consciousness will be altered.

Even then we don't know if our particular flavor of consciousness is somehow tied to organic matter. It's possible that when you transfer your consciousness to electronics, it'll really only simulates conscious - but not have any.

While I do think that most likely the brain and consciousness could be mapped onto a computer, essentially programmed, it's still not entirely out of the question that this perhaps that isn't actually true.

If consciousness is somehow tied, either by way of physics or some emergent property, to organic matter, then your consciousness would disappear - perhaps replaced by some new kind of consciousness that can only exist in electronics or no consciousness at all.
helios wrote:
I.e. it deconstructs your body and reconstructs it someplace else. Such a process would kill you every time.
I don't think that dilemma exists because i dont think that atoms are relevant for your mind. The mind/consciousness is not tied to the atoms.

What i believe is the following:

The whole life is about renewing cells. The body gets rid of old cell and create new ones. Therefore we eat and drink. The atoms forms the cells but it is not relevant what atoms (okay the type is kind of relevant).
The DNA is basically a todo list for building a new cell. Our memory is stored within this configuration of cells.

So when you are able to exactly copy the human body you create a twin. In that moment when the twin has no further experiences [s]he is you and it's no problem to dismiss the old atoms.

Dying is something else. It is the process when the cells dissolve.

You are your brain.
I don't think so. The brain is certainly vital to control things, but it is not completely you.
Last edited on
Just as if you were inheriting from a class. The components you inherit and final outcome is still bound by physics. However, that doesn't mean they're reducible to physics.
The combination of subatomic particules and of atomic into molecules is more like composition than inheritance.

I stumbled upon Hempel's dilemma while doing some research. Basically, it states that, when claiming everything can be reduced to physics, that ]must mean they think either everything is reducible to the physics that we have now (which is incomplete) or some "future" complete physics (which is unknown and would cause circular reasoning -> things are reducible to physics because future physics says so).
This implies a misunderstanding on the part of the person who posits the dilemma. Science assumes that the universe works on consistent principles. Initially we don't know what those principles are, so we design experiments to learn about them. With sufficient information, we create mathematical models that attempt to predict or approximate the behavior of the universe. Examples of such models are the laws of thermodynamics, relativity, quantum mechanics, etc. When someone says that all phenomena are reducible to "physics" they don't talk about any particular model of the universe, but about the actual principles driving the universe.

https://www.talkorigins.org/faqs/evolphil/reduction.html
You know, it's weird. I've never heard a chemist complain about reductionism. Biologists say shit like "genetics is not reducible to physics" and then wonder why nobody takes them seriously.

Many Mendelian genes are made from many DNA molecules, and many populational traits are coded for by many Mendelian genes. Simple reduction will not work. What Williams [1966] called an 'evolutionary gene' is just a unit of heredity that is 'visible' to selection, and it could be an entity at any level - a molecule, a Mendelian gene or even a populational trait.
This, again, is a misunderstanding of the situation. This is what happens when you forget that you're understanding the world through abstractions.
In biology, a gene is a basic unit of heredity and a sequence of nucleotides in DNA that encodes the synthesis of a gene product, either RNA or protein.
"Gene" is an abstract concept that relates to abstract concepts. A subatomic particle doesn't know about genes or even molecules. It will just do what it does, whether it happens to be part of a nucleobase or a protein.

Odd idea... popped into my mind. We sometimes like to assume something like X is true if we had infinite computing power and memory available. If something truly takes infinite amounts of computing power to be true, that would practically, and perhaps then literally and theoretically, be impossible to be true.
No, does not follow. Gödel assures as that there are true statements that cannot be proven to be true, even with infinite resources. Given finite resources, the number of true statements that can be proven is obviously smaller.

The thought experiment of Laplace's demon doesn't require infinite resources. It just requires as many resources as the universe runs on (whatever those may be) and perfect knowledge. Oh, and a new universe to place the demon in, since you can't run a simulation of the universe that the simulation runs on, obviously. To simulate the Earth, in particular, you would only need to simulate the observable universe, since everything outside it is causally disconnected from the Earth.

You mean transform the atoms that make your brain from organic matter to electronics?!
I meant at the cellular level. The nanomachines or perhaps even micromachines would replace your neurons with an equivalent electronic component. Obviously we would need to understand the nervous system enough to be able to produce not just electronics that can replicate the behavior of neurons, but also to interface with them.

It's possible that when you transfer your consciousness to electronics, it'll really only simulates conscious - but not have any.
An electronic neuron good enough to fully replicate a biological one would necessarily mesh into the system seamlessly, without disrupting any of its processes.

If consciousness is somehow tied, either by way of physics or some emergent property, to organic matter, then your consciousness would disappear - perhaps replaced by some new kind of consciousness that can only exist in electronics or no consciousness at all.
At worst it would be no different than growing from childhood to adulthood. Your personality and your perspective might change, but this happens throughout life either way. The illusion of continuous existence would be preserved.

So when you are able to exactly copy the human body you create a twin. In that moment when the twin has no further experiences [s]he is you and it's no problem to dismiss the old atoms.
If you create an exact duplicate of yourself, down to the positions of all your particles, do you inhabit two bodies, or is there simply a new being with your memories who behaves exactly as you would? If I then take all your atoms and separate and sort them into cubes by their elements, from your perspective, what has happened to you? Are you alive in your subatomic-level copy, or have I killed you?
I meant at the cellular level

If you leave atoms behind, you could theoretically create a completely different consciousness. But it seems possible if the process is actually plausible.

The illusion of continuous existence would be preserved.

Perhaps, but the consciousness being one is still one in the same is important. Conscious may evolve over time, but it's not really an illusion if is the same cells and atoms producing that evolving consciousness - you would be actually continuously existing, just also changing.

If you create an exact duplicate of yourself, down to the positions of all your particles, do you inhabit two bodies

Even if possible to do so, no. There would also have to be a connection between the two brains. If they could communicate the same way one brain does, then somewhat. You'd probably have a melding of two consciousnesses - which I won't even try to predict what that would be like. Maybe two people talking at the same time or one gigachad consciousness.

If I then take all your atoms and separate and sort them into cubes by their elements, from your perspective, what has happened to you? Are you alive in your subatomic-level copy, or have I killed you?

You've killed me! Not only are the atoms important, but the structures they create. Your specific brain anatomy and genes are just as important.



When someone says that all phenomena are reducible to "physics" they don't talk about any particular model of the universe, but about the actual principles driving the universe.

Then any higher explanations, even if by one definition or another is "reducible" to physics, it wouldn't mean that there couldn't be room for free will.

This definition of physics is broad. Even if behaviors that are impossible to predict through seeing atoms and forces alone occurred, you'd just incorporate this new behavior into physics.

If there is in fact a way for conscious minds to influence our brains in a way that constitutes free will, physicists will say, "Aha! Well that's all from physics anyway". They're not wrong, everything does obey physics, but it still doesn't prove that everything is reducible to physics.

When talking about atoms and forces, it may make sense to say you can theoretically predict every interaction every atom will ever have. But then you may find out this doesn't actually work. This can be way of what I said earlier, needing infinite resources could simply mean the theory itself is flawed, or through quantum mechanics finding that electrons are actually inherently unpredictable.

We already know that our conscious minds influence our thoughts and actions - otherwise we wouldn't even be talking about it. The only question becomes whether our thoughts are direct byproducts of physics in a way that does not allow the brain to have free will, or if there exists an emergent property built from physics but only explainable at a higher level than physics.


Gödel assures as that there are true statements that cannot be proven to be true, even with infinite resources.

I don't think that applies. Those proofs aren't trying to rely on infinite resources, they are instead said to be unprovable to begin with. We can have this equation pretty much:

Theory + n*resources != Provable

So you can rewrite to show that EVEN with infinite resources it's not possible:

Theory + Infinity*Resources != Provable

However, a proof that requires infinite resources to be theoretically viable would be vastly different.

It just requires as many resources as the universe runs on (whatever those may be)

Space is also a resource. You'd only have to allow for infinite amounts of space in order to say that such a simulation would require infinite amounts of resources. This is not to mention what else could be infinite.

To simulate the Earth, in particular, you would only need to simulate the observable universe, since everything outside it is causally disconnected from the Earth.

That's not entirely true. Everything in OUR observable universe has its OWN observable universe. Things at the edge of our observable universe could be affected by things just beyond our reach but within theirs. Then we can be indirectly affected by the unobservable universe.

It's also not entirely impossible that we could be directly influenced. You could have particles that have been tangled and then separated, one to our observable universe and one outside. These particles would communicate faster than light and may bridge the gap between the observable and unobservable universe.

Hidden variables is proven false, so there's no getting out of that one.

Biologists say shit like "genetics is not reducible to physics" and then wonder why nobody takes them seriously.

Perhaps it's not? Neither of us are biologists, and they implicate that there is a many-to-many relationship with genetics. This means explanations can only occur at higher levels and reduction will be impossible - even though biology is built upon physics.

A physicalist's view would be that through physics, physics gained self-awareness. Quantum mechanics has an indeterministic view point of the world thus far - and quantum mechanics is how we understand the big bang - which would be the reason for everything in a deterministic viewpoint.

Classical physics has come short time after time in the realm of the very tiny - yet it is only through the principles of classical physics that you can get a deterministic viewpoint of the world. Quantum mechanics throws a wrench at deterministic viewpoints.


Again, we're not biologists, and from what I keep reading they have reason to believe that by only looking at the sum of parts, you could never adequately explain or predict the whole.

They claim some parts of biology are not only not reducible to physics, but not even to chemistry.

Alex Rosenberg and others have written interesting things about it. Though, I wouldn't know for myself how valid such arguments are.


However, again, if the reductionist view point insists that there is no practicality in using reductionism (since atoms and forces won't be adequate for us to use for explaining disease) and that everything which arises from atoms and forces will also then also be explainable through atoms and forces - then you've taken a rather pointless position.

Consciousness seems inconceivable to come about from physics - yet it has. The reductionist will say, "sure, but every part of the brain works on biology which is just chemistry which is just physics". But then denies that there could be the possibility of free will.

If physics is predictable on the scale of atoms - couldn't emergent properties make it so that predictability goes away?

The reductionist view is so regressive that there's almost nothing to argue. It's unfalsifiable at this point - since anything that ever comes about will be said to just be physics. Nothing less than magic would prove it wrong.

You might then say, "Free will would be magic! That would disprove physicalism!" But that's not the same thing. Evolution is falsifiable because there is something non-magical we could find that would bring our view into question. That's not the case with physicalism.
I don't think that dilemma exists because i dont think that atoms are relevant for your mind... The brain is certainly vital to control things, but it is not completely you.

Then you're going into spooky ghost town. The brain is completely you. The atoms that make your brain, if all replaced, would kill you and replace you.

If you recombine the old atoms into the same brain it was replaced with, which would be the original you? Obviously the reconstructed brain with the old atoms.

Our memory is stored within this configuration of cells.

Yes, but where is our consciousness? It doesn't matter if there is "a" consciousness, what matters is that it's the same consciousness.

If you were able to copy your memories, DNA, and every other relevant thing into a new brain, you're leaving yourself behind in the old brain - that's where your original consciousness is.

All you've created is a copy that'll have a new consciousness but will never know that it's only a copy - the new person will feel like the old person.
There's no way to program understanding or purpose.
I wonder how a mundane airconditioner controller would feel about that.

It's whole existence revolves around it's purpose in understanding the temperature of the room it's operating in, all the result of hardware driven by a program plus energy.

The saving grace is probably that the program doesn't extend to emotional responses. Just hook it up to Siri and if she gets upset the power gets shut off or it short circuits and burns the house down.
but it's not really an illusion if is the same cells and atoms
The atoms and cells in your body, including the ones in your brain, are continuously being replaced.

you would be actually continuously existing, just also changing.
Which is exactly what we observe. People change as they age.

They're not wrong, everything does obey physics, but it still doesn't prove that everything is reducible to physics.
These two statements are in direct contradiction.

When talking about atoms and forces, it may make sense to say you can theoretically predict every interaction every atom will ever have. But then you may find out this doesn't actually work.
If it can't be done even in principle then the only conclusion left is that the universe doesn't follow consistent laws. It cannot be investigated through physics because there's a trickster god constantly changing the rules to fool us.

Everything in OUR observable universe has its OWN observable universe. Things at the edge of our observable universe could be affected by things just beyond our reach but within theirs. Then we can be indirectly affected by the unobservable universe.
Haha, I knew you would say that. The observable universe is a four-dimensional cone that encompasses the entirety of space-time that can causally affect Earth. Because of the speed of light, something currently outside the observable universe cannot possibly affect Earth in the future, even indirectly by having an effect on an object close to the edge and that effect being visible by us. The light from an object close to the edge of the observable universe that could be affected by something outside would take so long to reach us that it would recede outside the observable universe before those effects could be seen here.

You could have particles that have been tangled and then separated, one to our observable universe and one outside. These particles would communicate faster than light and may bridge the gap between the observable and unobservable universe.
Quantum entanglement does not transmit information faster than light.

Neither of us are biologists, and they implicate that there is a many-to-many relationship with genetics. This means explanations can only occur at higher levels and reduction will be impossible - even though biology is built upon physics.
Like I said, that irreducibility only exists because they're dealing with abstractions.

Imagine that you have a bunch of computers that you want to use as if they were one big hard drive. You write software that presents a contiguous block device but actually distributes the data through the network while also handling redundancy and node disconnections. Now you add a file to this virtual unit. In which computer is the file? That would be difficult to say, right? The file may be broken up in pieces and each of those pieces may exist in multiple computers. Working at the level of "file" the best we can say is that the file exists in the network, because we're several levels of abstraction deep. However, if I instead ask in which computers byte offset 42 is stored, that's much easier. You just turn off computers until it can no longer be retrieved, and then turn them back on then off one at a time, writing down when the byte is accessible.

Another example. In the general case, it's very difficult to correlate a C++ line number to a sequence of bytes in an optimized binary. The source is an abstraction of the program's behavior, and the compiler may perform radical rewrites of the code as long as that abstraction is preserved, such that a particular line may be reordered, intertwined with other lines, or even completely absent in the final executable. That doesn't mean that the program's behavior is emergent and irreducible to the execution of a sequence of programmed instructions.
(Or, to be even less abstract, to quantum interactions in semiconductors.)

Again, we're not biologists, and from what I keep reading they have reason to believe that by only looking at the sum of parts, you could never adequately explain or predict the whole.
There are two reasons, which I've already mentioned:
1. We don't yet have a theory of everything.
2. Even if we did, to study a living being using physics one would need to deal with so much information that it could never be done. We use abstractions because they allow us to keep the amount of information in check.

They claim some parts of biology are not only not reducible to physics, but not even to chemistry.
Anyone who says that is a religious moron.

However, again, if the reductionist view point insists that there is no practicality in using reductionism (since atoms and forces won't be adequate for us to use for explaining disease) and that everything which arises from atoms and forces will also then also be explainable through atoms and forces - then you've taken a rather pointless position.
Reductionism and emergentism are philosophical positions. They're inherently pointless.

The reductionist will say, "sure, but every part of the brain works on biology which is just chemistry which is just physics". But then denies that there could be the possibility of free will.
Are you implying there's a contradiction? Because I don't see it.

If physics is predictable on the scale of atoms - couldn't emergent properties make it so that predictability goes away?
I've never seen an emergent property demonstrated.
The atoms and cells in your body, including the ones in your brain, are continuously being replaced

Your brain cells definitely aren't being changed - those are the same cells from birth. Only the atoms that make up the cells may be changed with blood flow giving nutrients and taking waste. However, I highly doubt all the atoms get replaced, some must be too central to come and go.

These two statements are in direct contradiction

They're not.

If it can't be done even in principle then the only conclusion left is that the universe doesn't follow consistent laws

That could be possible. We're seeing that with Heisenberg's Uncertainty Principle. Let's not forget things going faster than light unlike Einstein's predictions. We still haven't solved many important questions.

It cannot be investigated through physics because there's a trickster god constantly changing the rules to fool us

The multiverse theory would make it relatively impossible to be deterministic if our universe is receiving any kind of energy from a universe with a different set of physics.

even indirectly by having an effect on an object close to the edge and that effect being visible by us

Why not? Light could reach a certain planet that will use that energy. The light will never reach Earth, but it could cause a chain reaction that eventually does. It only needs to be possible for you to have to take into account everything.

Quantum entanglement does not transmit information faster than light

Says who? Ghost busters?

irreducibility only exists because they're dealing with abstractions

It's funny that one would consider the higher level analysis to be the abstraction.

Still, the analysis at higher levels, if true, means physics must obey those rules when building those adequately complicated systems.

There is no pragmatic way to insist on how the laws of physics can give birth to consciousness. If physics cannot create a hierarchy of laws and effects that bound its creations but not reducible to the parts, then how can there be a physical explanation for consciousness?

How can you explain in physical terms how physics became aware of itself? You can't confidently say it is just a computational barrier without more information.


Are you implying there's a contradiction

The contradiction is that you don't know how physics can give birth to consciousness yet assume it cannot then give birth to free will.


Here's a thought: If our desires came from some external place - outside the universe, and the processing our brains happened some place where energy flowed in and out truly unpredictably. Would we then have free will?

Is free will unattainable in your view point because of determinism? Or because the laws of logic and biology could not possibly create such a thing?


Physics is determinable. Through evolution, you generate brains that are capable of decision making. Brains are systems.

Your brain receives stimulus and gives some sort of output. Physics determinism says that the stimulus will make you think, "I'm going to get chocolate ice cream".

Your brain, "free will", sees the information, processes it, and then makes a decision. This process is what we call free will - simply because our brains give rise to conscious thoughts which we use to make a choice.

If physics is determinable, does that mean your computer only gives the illusion of running your code? No, it actually runs your code, that doesn't change just because the system runs on physics and is determinable.

Hence, even if your brain was entirely determinable (which we've yet to see), it doesn't mean that free will is absent. It only means that the system of free will can be predicted.

Your free will wouldn't "happen" to coincide with the predictions of physics. Rather, your free will would be a predictable system, and the input given to it is predictable (theoretically) - hence one's future decisions are predictable.

If the moon gained consciousness and decided it was happy orbiting the Earth, why would it stop? If its within its power to stop - and it had the desire to stop, why wouldn't it?
zapshe wrote:
The brain is completely you
Then let me ask a question: Where does the brain end? When you cut your finger the brain is informed and reacts. So your finger part of the brain or not?

Yes, but where is our consciousness?
Let me rephrase it a bit:
Is the consciousness tied to the atoms?
Then when a certain construct comes into existence it builds a consciousness. This would also mean when this construct changes the consciousness would also change. Thinkable.

Or

Are the atoms tied to consciousness?
Then consciousness would exists and at one point it gathers atoms and builds a body (with the help of the parents(?)). Then this consciousness would be able to change its construct within some given limits.

helios wrote:
Are you alive in your subatomic-level copy, or have I killed you?
Well, for me [as the copy] everything is fine. I'm breathing, my heart is beating and I can do everything I could do without the incidence. So yes I'm alive, just at another location.

For me the question would be: When two me exists who are identical, is it okay to dump one.

But actually I have no concept of a 'subatomic-level' and what it means there...

helios wrote:
If you create an exact duplicate of yourself, down to the positions of all your particles, do you inhabit two bodies, or is there simply a new being with your memories who behaves exactly as you would?
Yes, that's tricky. That boils down to: Is it possible to duplicate a consciousness or not?
Where does the brain end? When you cut your finger the brain is informed and reacts. So your finger part of the brain or not?

Your nerves communicate to your brain. That's like asking, "Where does your computer end? When you plug something into a port, it detects it." But the processing part of the computer is the CPU (and GPU technically).

Your brain doesn't extend to your fingers, your brain has nerves, much like the linings through the motherboard to connect various circuits. They don't "extend" the CPU, it connects the CPU.

Is the consciousness tied to the atoms?

If you make 2 completely identical CPUs, what's to separate them other than that they are made of different atoms?

Then when a certain construct comes into existence it builds a consciousness. This would also mean when this construct changes the consciousness would also change. Thinkable

Well, yes. If you sustain brain damage, your consciousness is altered.

Then consciousness would exists and at one point it gathers atoms and builds a body (with the help of the parents(?)). Then this consciousness would be able to change its construct within some given limits.

This one makes no sense.
Last edited on
Your brain cells definitely aren't being changed - those are the same cells from birth. Only the atoms that make up the cells may be changed with blood flow giving nutrients and taking waste. However, I highly doubt all the atoms get replaced, some must be too central to come and go.
https://bebrainfit.com/brain-cells-regenerate/

Why not? Light could reach a certain planet that will use that energy. The light will never reach Earth, but it could cause a chain reaction that eventually does. It only needs to be possible for you to have to take into account everything.
Because the speed of light is an absolute limit on the transmission of any information, not just light. If light hits a remote object causing some effect on that object, and we're able to observe the effect on that object, then the information from the original light is being transmitted to us, thus the original light should also be visible (if unobstructed) If it was possible to send information indirectly from A to C by passing through B, where A and C are on opposite sides of an event horizon relative to each other, but neither A relative to B nor B relative to C are, then it would be possible to send information faster than light.

Says who? Ghost busters?
Says... the Wikipedia article on quantum entanglement, and every other source I've read on the subject?
Simply put, quantum entanglement causes (for example) the quantum spins of two entangled particles to replicate each other seemingly at infinite speed, because you can observe the spin on one particle and the spin is instantly replicated on the other one. This has been tested by carefully timing measurements between distant entangled particles. This doesn't transmit information faster than light, though, because you can't control the spin of either particle in such a way that it transmits across the tangle, you can only observe it. Once you've observed one spin, the spin of the other particle is locked in and won't change, so no transmission is possible in either direction. If you were to change the spin (I'd guess) the particles would detangle.

The quantum world is weird, man. Look up "delayed-choice quantum eraser". That experiment leads me to believe all current interpretations of QM are wrong.

Still, the analysis at higher levels, if true, means physics must obey those rules when building those adequately complicated systems.
Since they're abstractions created by humans derived empirically and not from first principles, no, it does not mean that. For example, physics does not obey Newtonian mechanics at very high speeds, nor at very small scales.

If physics cannot create a hierarchy of laws and effects that bound its creations but not reducible to the parts, then how can there be a physical explanation for consciousness?
I'm not sure I understand the question. "How can there be a physical explanation for consciousness"? The facts are:
1. We don't know how or why consciousness arises. I.e. what's the bare minimum required in order for an inanimate object to have subjective experience?
2. Consciousness is a process based on the brain. Damaging the brain damages the consciousness.
3. The brain is made of matter, which is bound by the laws of physics.
I don't see anything here that precludes consciousness from being explainable by a large number of interactions between particles, we just don't know what that might be. What prompts the question "how can there be a physical explanation for consciousness"?

How can you explain in physical terms how physics became aware of itself? You can't confidently say it is just a computational barrier without more information.
Actually, what I said is that it's not feasible to study biology at the subatomic level. How many elementary particles are there in a virus? Trillions?
To explain consciousness you don't need to do it at the subatomic level, just like how you don't need to hook up an oscilloscope to a CPU to debug your programs. You just need to figure out the structure of the brain to a sufficient level of detail.

The contradiction is that you don't know how physics can give birth to consciousness yet assume it cannot then give birth to free will.
Because free will contradicts determinism and causality, which are basic assumptions of our understanding of the universe, but assuming them does not by itself allow us to deduce an explanation of consciousness. I think assuming determinism rather than free will is the parsimonious position.

Here's a thought: If our desires came from some external place - outside the universe, and the processing our brains happened some place where energy flowed in and out truly unpredictably. Would we then have free will?
In that case, our will would be non-deterministic but not necessarily free, since our minds might be bound by the rules of that external place.

Your brain, "free will", sees the information, processes it, and then makes a decision. This process is what we call free will - simply because our brains give rise to conscious thoughts which we use to make a choice.

If physics is determinable, does that mean your computer only gives the illusion of running your code? No, it actually runs your code, that doesn't change just because the system runs on physics and is determinable.
I've already addressed this. By this interpretation, your will is no more free than that of a computer or a clock, or even a rock. A rock "decides" to stay firmly on the ground using the exact same mechanisms that you "decide" to be unconvinced by my arguments. Neither ever had the option to do anything else.

Hence, even if your brain was entirely determinable (which we've yet to see), it doesn't mean that free will is absent. It only means that the system of free will can be predicted.

Your free will wouldn't "happen" to coincide with the predictions of physics. Rather, your free will would be a predictable system, and the input given to it is predictable (theoretically) - hence one's future decisions are predictable.
I've been trying to figure out how to respond to this for the past ten minutes.
You say that (under the supposition that the brain is reducible to physics) it's not that your decisions coincidentally match the predictions of physics. That must mean that what you call your free will could hypothetically be caused purely by physics, right? Okay, then I must ask: in what sense of the word is your will "free"?
Secondary questions: If I accept that free will in this sense exists and that people have it, then what else has it? What's the least complex thing that has free will?
Last edited on
Pages: 123