Does Neuroscience Challenge the Existence of Free Will?

[Ferris_bg]

Hi Ken G,

The topic of http://www.iep.utm.edu/freewill/" . The theory wants to mark causal interaction as possible, but could it be? Can we have downward causation, when we speak about token identity theories? Can we have even any mental causation?

If we want a mental event M1 to cause a physical event P2 and if we want the causal status of the mental to derive from the causal status of its physical realizer P1 (so that the theory doesn't fall in the substance dualist category) we are faced with over-determination (P2 could be realized by M1, as well as by P1 alone). If there are no greater causal powers that magically emerge at the higher level of M1 (if we want the theory to stay a materialistic one) then the causal powers of M1 are identical to the causal powers of P1, which means that P1 is the only realizer of P2, thus M1 becomes epiphenomenal. You can read more about this http://www.iep.utm.edu/mult-rea/#H4".

So, in the materialistic view you can either have mental causation identical with the physical causation or you can embrace epiphenomenalism and qualia. In both ways free will is impossible. If you want to find free will, you must seek it outside the materialistic domain.

Q_Goest,

In your post https://www.physicsforums.com/showpost.php?p=3179362&postcount=90" you say you don't believe in the phenomenal-physical correlation and basically you reject epiphenomenalism. And at first it doesn't looks logic, how can one make a knowledge claim about consciousness if it's epiphenomenal? But does the agent's association of the conscious experience of some event and its labeled state in the brain contradict in any way? The definition of the word "consciousness" in the brain state is not associated with the experience of it, but does this interfere the brain to be able to label certain physical state? Think about it, how will you explain the word "consciousness" to a little boy and what association does his brain make. For me epiphenomenalism implies that in exactly every millisecond your brain takes the optimal decision based on the available information. Even when you do something anti-evolutionary (take a lot of drugs, commit a suicide) it must be somehow justified in your brain calculations. Because if it's not, epiphenomenalism is wrong (remember you don't have taken the drug because YOU liked it, but because your BRAIN liked it).

 

[Pythagorean]

...

You can go back farther than computers. The computer is basically just a glorified calculator used to solve differential equations that we can't solve by hand (or we could... but it would take hours and pages where the computer can do it in seconds and kilobytes).

But think about this: let's say you have some giant system of N differential equations to describe the whole universe. You have every single interaction reduced to a handful of variables. Now all you need to do is put in your initial conditions for those variables.

What do you do? Your theory already accounts for everything in the universe, yet your theory doesn't account for how the initial conditions arose. Do you make the initial conditions a function of some part of the system? So now there was always this loop and never a beginning or end? I'm puzzled, personally, I have no idea what I'd do.

Anyway, I'm hoping this demonstrates that the science and the philosophy are completely different, just like models and reality. As another examples, we know that quantum mechanics underlies all classical observations, yet we naively model things in the old classical view. Why? Because it's effective, it's productive, it works. This is not the same way I approach the problem in a philosophical setting.

 

[nismaratwork]

You can go back farther than computers. The computer is basically just a glorified calculator used to solve differential equations that we can't solve by hand (or we could... but it would take hours and pages where the computer can do it in seconds and kilobytes).

But think about this: let's say you have some giant system of N differential equations to describe the whole universe. You have every single interaction reduced to a handful of variables. Now all you need to do is put in your initial conditions for those variables.

What do you do? Your theory already accounts for everything in the universe, yet your theory doesn't account for how the initial conditions arose. Do you make the initial conditions a function of some part of the system? So now there was always this loop and never a beginning or end? I'm puzzled, personally, I have no idea what I'd do.

Anyway, I'm hoping this demonstrates that the science and the philosophy are completely different, just like models and reality. As another examples, we know that quantum mechanics underlies all classical observations, yet we naively model things in the old classical view. Why? Because it's effective, it's productive, it works. This is not the same way I approach the problem in a philosophical setting.

A "shut up and calculate" philsopher? If you weren't so clearly a dude, I'd be in love.
(not sarcasm)

 

[apeiron]

And yet, it is really many-worlds which is holistic, and has a whole that is more than the sum of its parts.

I would still argue not as many worlds is exactly the sum of its parts. Every locally forking history accumulates without any constraint. Holism would require that the local freedom to branch would be restricted so the system only manifested some paths and not all of them.

By contrast in QM views of causality, a Feynman sum over histories approach to collapse is holistic as all paths "happen" but then there is a global constraint to some single self-consistent event.

 

[apeiron]

Personally, I always get a chuckle when I see the term "fundamental" used in physics. What does that even mean? Perhaps when we look at the history of physics, we should start relating to what physics actually is rather than how we might like to imagine it. The natural conclusion is that the word "fundamental" by itself does not have meaning in physics, but "more fundamental" does. Given this, we should not be surprised that causality is not fundamental, but we can perhaps view it as "more fundamental" than a concept like space. The prevailing question of this thread is then, "which is more fundamental, causality or free will?" Or perhaps neither emerges from the other, but both emerge from something else.

Causality is of course fundamental and freewill as near epiphenomenal as you can get . Causality would be our general or universal model of why anything happens (why even existence happens), while freewill is just some vanishingly rare, relatively impotent on the cosmic scale, feature of a complex system.

In philosophy, the fundamental is the general. In physics, it is generally taken to be the smallest scale - which is why atomistic reductionism is the driving idea.

And when it comes to identify these general principles or universals, philosophy finds that they are always dichotomies or complementary/synergistic/asymmetric pairs.

So as well as the local, there is the global. As well as the discrete, there is the continuous. As well as flux, there is stasis. As well as chance, there is necessity, etc.

Which is why it is no surprise that causality itself is dualised. As well as bottom-up construction, there is top-down constraint.

 

[apeiron]

Let’s talk about what these models are for a minute because I think the philosophy of why they are the way they are is being overlooked.

It's great that you are willing to get into the details of a defence of your view. And my reply is that you are missing the wood for the trees .

What you are highlighting here is simply the fact that allowing a system to go to global equilbrium allows you then quite properly to drop the global causes from your model because now you are only interested in what can change - the local variables, the local fluctuations, the local events. This is what reductionist modelling is all about.

It is right there in Newton's three laws of motion. The first two laws atomised the notion of local action into a force and a mass. Mass could have intrinsic motion which was inertial, and that made any globally observable change in motion the result of an atomistic force (a force vector).

So already in the first two laws, Newton's great reductionist simplification was to equilibrate away the global spacetime backdrop. Taken the greek atomist's notion of the void, he said the background exists, but it is causally inert. It is simply an equilbrated or unchanging stage upon which there is a localised play of atoms - atoms of mass and atoms of force.

Then to make this highly reduced view of reality fly, he had to introduce his third law of action~reaction. For every forceful action, there is an equal and opposite forceful reaction - a little matching localised anti-vector.

Patently the reaction vector is not actually a symmetric entity. Instead it sums up all the contextual constraints that are found to impinge on the locale. If you push against the wall, then it is not just several square inches of wall that pushes back. It is the building, the planet to which it is attached, the gravity fields which affect the planet, etc.

The third law is the local equilbrium correction! The first two laws removed the generalised background and the third quietly accounts for any disturbances of the global state by localising it to another linear and atomistic event - a reaction vector.

So this is the "philosophy" of physics - or at least the highly successful modelling strategy on which all mechanical thinking is based. Equilibrate away the global causes, the context that constrains, and you can then just describe reality in terms of local atomistic entities and local forceful changes. Just treat reality as a collection of actions happening in a mute void.

Now FEA just repeats the same exercise. If you can't equilibrate away the whole global story at once, then break the job up into a suitably grained set of compartments. Create localised equilibration stories that add up with suitably low error to give you a globally equilibrated model.

Does this then say that global downward acting constraints don't exist? Or that reductionist modelling finds ever more clever ways around them?

Now this thread was about the neurology of freewill. (Not modelling neurons with FEA).

The kind of systems that FEA is suitable for modelling is stuff like fluid dynamics. This is the non-living world where global constraints are holonomic. We are safe to presume the constraints or boundary conditions are at equilbrium and unchanging. Locally the aircraft wing may be subject to some complexity due to emergent turbulent features. But generally temperature, pressure, material strengths, viscosity, are a stable backdrop to the model.

There is not a local~global interaction so that for example the flex of the wing causes a tropical storm that sends a bolt of lightning that changes the material strength of the wing, or even just causes a dramatic pressure drop in the vicinity of the wing. No, the FEA analysis rules out interactions across scale by choice.

But for living systems, we are now talking about systems that have non-holonomic constraints. They do have the informational machinery (such as genes, words, membranes, action potentials, etc) to control their own boundary conditions or downwards acting constraints.

So to model living systems, we have to model that ability to change the global constraints - for meaningful reasons. Which is why I keep challenging you to reply to the literature on top-down selective attention and its power to reshape local neural receptive fields.

You would rather keep the discussion focused on the most reductionist models of single neurons that you can find. And yes, you can take what a receptor pore does and model it as an isolated mechanical device sitting in a stable equilibrium world utterly unlike the real world of a receptor pore. It will tell you something about the local degrees of freedom that the device might have. But it cannot then tell you anything about the kinds of global constraints that act on those degrees of freedom. You literally cannot in principle see them.

Now you can do a Blue Brain exercise and throw a lot of devices together and simulate - see what kind of global organisation arises to constrain a network of artificial neurons. If you have built your simulation with local components that can change their behaviour (as is familiar with neural nets with nodes that can adapt their local weights), then you can start to get a realistic development of local~global interactions.

But a simulation is NOT a model. The results you are celebrating are the observable output, not an axiomatic input. You are demonstrating an effect, not a cause.

A proper model in this context would be one where you have a handle on both the bottom-up and top-down sources of causality and so can compute the outcomes directly - predict the observable state rather than merely discover it post-hoc.

So this is why systems modelling is different from reductionist modelling. Reductionism wants to deal only in local causation (and so finds ways to equilibrate away any global effects to make them a "void" - an unchanging backdrop). Systems modelling recognises that global constraints can be an active part of the mix and so seeks to include them in the model.

This is of course very difficult to do as yet. In fact it could be another 20 to 30 years before we have the real breakthroughs in this area. Everyone thought fractals, chaos theory and non-linear dynamics was some kind of mathematical modelling revolution. But that was just a first ripple of the change that could come.

 

[Ken G]

I would still argue not as many worlds is exactly the sum of its parts. Every locally forking history accumulates without any constraint. Holism would require that the local freedom to branch would be restricted so the system only manifested some paths and not all of them.

It appears we have a different idea of the meaning of "holism" as it pertains to quantum mechanics, so this might be interesting to delve into. To me, the quintessential example of holism is the violation of the Bell inequality. This requires correlations of a specific type, i.e., not the type of correlations we have classically-- instead, it requires a concept of a joint wave function, in effect. The simplest example there is a Bell state, like |A>|a> + |B>|b>, where capital letters are for one part of the system and small for the other, and A and B are two different outcomes. Classically, there's just three things there, A-like aspects, B-like aspects, and the probabilistic combination thereof, but the Bell state allows algebraic correlations that are also part of the state, so the state has four elements-- there's an extra element which is the algebraic consequences of the combination (the phase coherences), and that's what unifies the A-like and B-like aspects into a single whole, even when A and B are incompatible outcomes.

Seeing a macro reality as if it were akin to a Bell state is what many-worlds does-- Copenhagen treats the macro reality like the classical state, where the probabilistic element reflects our lack of information rather than something that is really true about the reality. In many worlds, the algebraic (holistic) combination is a legitimate aspect of the reality (it's one that we the observer are never privy to, because the phase coherences are of such a complex and intractable nature that we never see their consequences in any experimental outcome-- our degrees of freedom are limited to an "island" of coherent phases, the boundaries of which are unapproachable by the way we interact with our reality). In short, in Copenhagen, the observer effects are what determine reality, and in many-worlds, the observer effects are a prison that prevents us from seeing the reality. That's why I see the difference as which one subordinates to the other: the physics or the physicist. And the meaning of "holism" is also dependent on this choice-- if we subordinate the physicist to the physics, then what is "whole" is the mathematical concept, the state vector with all its invisible phase correlations that we cannot be affected by because they don't fit into our coherent subset. If we subordinate the physics to the physicist, then what is "whole" is the consistent history that physicist will use to describe their reality, and the "many worlds" seem fragmented.

By contrast in QM views of causality, a Feynman sum over histories approach to collapse is holistic as all paths "happen" but then there is a global constraint to some single self-consistent event.

Yes, you take the Bohr view that the observer effects are global constraints that reality must satisfy-- many worlds drops that requirement. I prefer the pragmatism of the Bohr perspective, but I think that when we take the pragmatic approach, we don't really have a holistic perspective left, because we cannot say that only a single "state of reality" is consistent with the global constraints-- there may be many states that are equally consistent, at the level of precision by which those constraints can be defined.

 

[Maui]

Causality is of course fundamental and freewill as near epiphenomenal as you can get . Causality would be our general or universal model of why anything happens (why even existence happens), while freewill is just some vanishingly rare, relatively impotent on the cosmic scale, feature of a complex system.

Your certainty isn't warranted.

It doesn't really matter how convincingly well the 4 fields are able to mimic the existence of solid objects with causal relations between them. They are not. And this is a truly tremendous point for philosophy, as by far the most consistent model of reality we have today(it's also the only consistent one) is that of fields and the field intensity that represents the probability that some classically looking(causality preserving) event will take place somewhere. If you think you understand the reality via your models- you don't. If you hope to understand the future TOE(the master equation) - you won't. Causality is just another part of the human baggage and will play a secondary role(or no apparent role) in a complex, self-consistent mathematical scheme(TOE).

Another ponit would be that if causality were so important and fundamental, we'd have figured out by now why the Schroedinger equation works as it does and probabilities would give way to certainty.

The universe isn't classical(this is certain), it's quantum and it looks classical under specific circumstances but in others, the classical universe worldview is totally inconsistent and can't explain a whol;e miriad of phenomena that the quantum worldview can. In the other thread in the qunatum forum someone is asking how two solid bodies can actually 'touch'. Go explain that in classical terms when the whole mindset that prompted the question is completely false.

 

[apeiron]

Your certainty isn't warranted.

But what I actually said was that even causality is something we "just model". Which is why both the reductionist model, and the systems model, could be "right" - each effective for their purposes, or within their domains.

 

[Ken G]

So, in the materialistic view you can either have mental causation identical with the physical causation or you can embrace epiphenomenalism and qualia. In both ways free will is impossible. If you want to find free will, you must seek it outside the materialistic domain.

Hello Ferris_bg, and welcome to the dialog. Personally I have no difficulty rejecting physicalism, it strikes me as weak logic. It seems to basically follow the path "because physical models have given us excellent predictive power, we will embrace the idea that the universe is physical." That is far from a syllogisim! Note also that when I say "physical" or "material", what I really mean (and what I would claim others really mean) is "physical models" or "material models", for the very use of the term invokes models. If one is not talking about models, then the terms "physical", "material", or "reality" have no distinctions, so add no content. Whatever is, is, and the labels we hang on it is of no consequence unless those labels characterize our models of it.

 

[Ken G]

Patently the reaction vector is not actually a symmetric entity. Instead it sums up all the contextual constraints that are found to impinge on the locale. If you push against the wall, then it is not just several square inches of wall that pushes back. It is the building, the planet to which it is attached, the gravity fields which affect the planet, etc.

I think you are making a lot of valid and insightful points, but I would see a need for correction in this one. Newton's third law is not that "the universe will conspire in a complex way to insure that every force results in a reaction force", it is that every single force that can ever happen comes in a pair, because forces are binary relationships, and it makes no difference what the rest of the universe is doing at the time. This is why it is not necessary for that square inch of the wall to be attached to anything in order for any force you put on it to meet with an equal and opposite force on your hand. If the wall is attached, it can remain stationary and provide that counter-force. If it is unattached, it will accelerate, and then its inertia will allow it to provide that counter-force. If the wall is unattached and has no inertia, then you cannot apply a force to it in the first place. This is why the ability to generate and experience forces is connected to inertia-- an inertialess charge, for example, would be singular in Newton's scheme, not because the universe would conspire in some complicated way to disallow it, but because that one single massless charge in the presence of an electric field yields a mathematical singularity all by itself within the Newtonian framework.

Indeed, it is often said that the third law is the reason for conservation of momentum, but when conservation of momentum is viewed as an axiom of translational symmetry, the logic flows in the opposite direction: the third law holds because of translational symmetry. So Newton's "reaction" concept is indeed a symmetry principle, as action without reaction would always have to break translational symmetry. That is what could only happen if the larger universe conspires to break that symmetry, so it is not the presence of the reaction, but its absence, that requires a larger universe to provide the necessary constraints.

 

[Ken G]

Causality is of course fundamental and freewill as near epiphenomenal as you can get . Causality would be our general or universal model of why anything happens (why even existence happens), while freewill is just some vanishingly rare, relatively impotent on the cosmic scale, feature of a complex system.

That is certainly a commonly adopted stance, but I would like to suggest another angle. I think that everything we think about how reality "works" is subordinated to how we interact with reality. And how we interact with reality is subordinated to how our brain works, including free will. So free will should not be expected to emerge from our study of reality, it should emerge (if at all) from our study of our brains.

So then the obvious conundrum emerges-- is our brain not a part of reality, so will not the same techniques that worked on "external" reality work on the brain? Maybe yes, and maybe no-- that's the point. The way we learn about reality is so caught up in the functioning of our brains that it is no longer obvious "which side of the microscope" the brain is on. We have no guarantee that the way we think about causation for, say, a charge in a field, will help us understand how the charges in our brains help us think-- it's just an article of faith that the kinds of questions that are pertinent will be the same.

Now, I have no other suggestions other than to apply the same techniques and ask the same types of questions, perhaps on a more sophisticated level (like the tennis match between up-down and down-up causation that you have been advocating), or on the reductionist level that has also been discussed. I'm just saying we should not start out with the assertion that this must lead us to the most fundamental results-- it may lead us down a dead-end street. A brain trying to understand a brain might be like a puppy chasing its tail, and when a puppy chases its tail, it has no idea why the tail keeps moving just out of reach every time the pupply makes a lunge at it, so it keeps on lunging, because lunging has worked so well on everything else.

In philosophy, the fundamental is the general. In physics, it is generally taken to be the smallest scale - which is why atomistic reductionism is the driving idea.

Yes, and it is remarkable that quantum mechanics obeys the correspondence principle-- there is nothing that appears to emerge in simple dynamical systems when passing from the quantum to the macro domain that invalidates the quantum analysis, it merely renders the quantum analysis inelegant. But as we both agree, that may be because of the way the problems are "rigged" to obey the correspondence principle from the outset-- the correspondence principle may not be a principle about reality, it may be a principle about physics, or how physics is generally done.

And when it comes to identify these general principles or universals, philosophy finds that they are always dichotomies or complementary/synergistic/asymmetric pairs.

So as well as the local, there is the global. As well as the discrete, there is the continuous. As well as flux, there is stasis. As well as chance, there is necessity, etc.

Which is why it is no surprise that causality itself is dualised. As well as bottom-up construction, there is top-down constraint.

The tennis match. Indeed I have long felt the "yin-yang" symbolism of eastern philosophy was one of the most profound concepts the human mind has ever developed-- the importance of both apparent contrast and deeper unity in generating understanding.

 

[apeiron]

I think you are making a lot of valid and insightful points, but I would see a need for correction in this one. Newton's third law is not that "the universe will conspire in a complex way to insure that every force results in a reaction force", it is that every single force that can ever happen comes in a pair, because forces are binary relationships, and it makes no difference what the rest of the universe is doing at the time. This is why it is not necessary for that square inch of the wall to be attached to anything in order for any force you put on it to meet with an equal and opposite force on your hand. If the wall is attached, it can remain stationary and provide that counter-force. If it is unattached, it will accelerate, and then its inertia will allow it to provide that counter-force. If the wall is unattached and has no inertia, then you cannot apply a force to it in the first place. This is why the ability to generate and experience forces is connected to inertia-- an inertialess charge, for example, would be singular in Newton's scheme, not because the universe would conspire in some complicated way to disallow it, but because that one single massless charge in the presence of an electric field yields a mathematical singularity all by itself within the Newtonian framework.

Not true. If the wall is not attached to anything, it just means you are having to exert less of an accelerative force and so the contextual back-reaction is matchingly less. All you have to overcome is the resistance of surrounding air molecules.

And it the wall "doesn't move", then the acceleration you exert will produce heat and noise. Equilibration is still happening.

If you and the wall are in space, then it becomes even clearer that who pushed who is a perfectly symmetrical question so far as the laws of physics are concerned.

I'm not sure what you are trying to say about inertia, but the whole point of an inertial body is that it is at equilibrium with the world. There is nothing acting on it to change its state of motion. There is no action, and so no re-action needed to re-equilibrate our local~global view of the body.

So Newton's "reaction" concept is indeed a symmetry principle, as action without reaction would always have to break translational symmetry. That is what could only happen if the larger universe conspires to break that symmetry, so it is not the presence of the reaction, but its absence, that requires a larger universe to provide the necessary constraints.

Err, it is the model that reduces the description of the world to a set of symmetries. That is how the actual entangled messy dynamism of the world can get abstracted away.

Science seeks the equilbrium stories because then it needs only measure the macrostate - the global constraints - and can ignore the confusion of local detail. The ontological mistake is then to call those robust and stable macrostates (like spacetime with its "inherent" symmetries) the fundamental ground of things.

 

[apeiron]

That is certainly a commonly adopted stance, but I would like to suggest another angle. I think that everything we think about how reality "works" is subordinated to how we interact with reality. And how we interact with reality is subordinated to how our brain works, including free will. So free will should not be expected to emerge from our study of reality, it should emerge (if at all) from our study of our brains.

What you might like as the way out of this conundrum are the philosophies offered by CS Peirce and Robert Rosen.

Peirce starts completely from "inside subjectivity" and works his way out to an objective description of reality in a reasoned fashion. It is very important that in logic he places abduction as prior to even induction and deduction.

And Rosen (a theoretical biologist who died about 10 years back) wrote about modelling relations theory. This is ultimately a theory of mind as the mind is a modelling system.

The tennis match. Indeed I have long felt the "yin-yang" symbolism of eastern philosophy was one of the most profound concepts the human mind has ever developed-- the importance of both apparent contrast and deeper unity in generating understanding.

A good book here is Joanna Macy's Mutual Causality in Buddhism and Systems Theory.

But I have to say that yin yang is a very undeveloped logic. The ancient greeks did this systems view of logic much better (see Anaximander, then Aristotle). Of course, it is an open question whether the greeks inspired the Taoists or the other way round as both views arose around the same time.

See for instance...http://arxiv.org/abs/physics/0309104

 

[nismaratwork]

What you might like as the way out of this conundrum are the philosophies offered by CS Peirce and Robert Rosen.

Peirce starts completely from "inside subjectivity" and works his way out to an objective description of reality in a reasoned fashion. It is very important that in logic he places abduction as prior to even induction and deduction.

And Rosen (a theoretical biologist who died about 10 years back) wrote about modelling relations theory. This is ultimately a theory of mind as the mind is a modelling system.



A good book here is Joanna Macy's Mutual Causality in Buddhism and Systems Theory.

But I have to say that yin yang is a very undeveloped logic. The ancient greeks did this systems view of logic much better (see Anaximander, then Aristotle). Of course, it is an open question whether the greeks inspired the Taoists or the other way round as both views arose around the same time.

See for instance...http://arxiv.org/abs/physics/0309104

Hmmm... good reading... do you have any more along these lines? I got a Kindle, and I'm in a book-buying mood... a wide range would be best.

 

[Ken G]

Not true. If the wall is not attached to anything, it just means you are having to exert less of an accelerative force and so the contextual back-reaction is matchingly less. All you have to overcome is the resistance of surrounding air molecules.

Not air resistance, inertia. And the force I can exert may have more to do with my physique than whether or not the wall is attached. My point is merely that no matter what the rest of the universe is doing, action/reaction is fundamental in Newton's system. The rest of the universe may get some input into how much force I can exert, that is a complex issue, but it doesn't get any say as to whether or not the force I can exert will be met with an equal reaction force, that is always true even in a universe of just me and the wall. This issue is actually a purely reductionist triumph-- if you analyze the force in terms of a bunch of pieces interacting with each other, and each of those pieces obeys Newton's third law (as is postulated in that system), then the whole will also.

I'm not sure what you are trying to say about inertia, but the whole point of an inertial body is that it is at equilibrium with the world.

Inertia just means mass, it doesn't mean not accelerating-- that's "inertial." Why those terms are used like that, I have no idea.

Err, it is the model that reduces the description of the world to a set of symmetries. That is how the actual entangled messy dynamism of the world can get abstracted away.

Yes, the model is an abstraction. It involves a background against which a universe can exist, and the background is translationally invariant. That means you could take that same universe and translate it, with no effect, as long as you translate everything. If you only translate part, then the "everything else" becomes a place you can put unbalanced forces and momenta, such that the part you are dealing with won't have an action/reaction principle. So if you don't see action/reaction working, then it implies (in this model system) that you are not dealing with the whole universe-- the presence of an "external" universe makes it presence known in the violation of Newton's third law, not in its enforcement.

Science seeks the equilbrium stories because then it needs only measure the macrostate - the global constraints - and can ignore the confusion of local detail. The ontological mistake is then to call those robust and stable macrostates (like spacetime with its "inherent" symmetries) the fundamental ground of things.

And indeed Newton's laws are found wanting for just this reason.

 

[nismaratwork]

@Ken G: I'd taking readng tips from you too...

 

[Ken G]

What you might like as the way out of this conundrum are the philosophies offered by CS Peirce and Robert Rosen.

Peirce starts completely from "inside subjectivity" and works his way out to an objective description of reality in a reasoned fashion. It is very important that in logic he places abduction as prior to even induction and deduction.

That does sound interesting. I must confess I had never even heard of "abduction" (other than the alien version) until I googled it, and it is indeed an important part of formal reasoning-- especially scientific reasoning. Indeed, I can see how easy it would be to argue that deduction and induction are just opposite extreme forms of abduction. This also gibes with the issue of truth vs. meaning that came up earlier: logic is often thought of as the arena for establishing syntactic truth, whereas experience is the arena of meaning. Induction and deduction are syntactic, a computer could be programmed to recognize them, but abduction would seem to straddle the domains of truth and meaning, sacrificing a formal stance in either realm in exchange for the ability to cross their boundaries. Fuzzy logic.

But I have to say that yin yang is a very undeveloped logic. The ancient greeks did this systems view of logic much better (see Anaximander, then Aristotle).

I'm not sure I would consider yin/yang a form of logic at all-- perhaps the Greeks took a different turn when they explored the power of logic. Indeed that may be the fundamental turn that distinguishes western vs. eastern thinking-- form vs. function, reason vs. introspection, consistency vs. contradiction. The Greeks gained great powers by banishing contradiction, and it has taken thousands of years to "play out the string" they started. But something might have gotten left behind, something that must someday be confronted in a theory of mind. Was yin/yang left behind for being underdeveloped, or just too far ahead of its time?

 

[Ken G]

@Ken G: I'd taking readng tips from you too...

I'm not as adept at tracking my sources of inspiration-- my thoughts come from a mish-mash of ideas I've been exposed to, including by people such as those on this thread. Probably the usual cast of characters in physics philosophy: Feynman, Wheeler, Wittgenstein, Penrose, Einstein, Bohr, Heisenberg, etc. Some of whom claim to be "shut up and calculate" types, a claim I never pay any attention to. :)

 

[nismaratwork]

I'm not as adept at tracking my sources of inspiration-- my thoughts come from a mish-mash of ideas I've been exposed to, including by people such as those on this thread. Probably the usual cast of characters in physics philosophy: Feynman, Wheeler, Wittgenstein, Penrose, Einstein, Bohr, Heisenberg, etc. Some of whom claim to be "shut up and calculate" types, a claim I never pay any attention to. :)

I don't mind, although it seems we're already fans of similar authors. Thanks very much Ken G, and if you think of anything later, just drop me a PM... I'm always hunting for reading material.

 

[apeiron]

Was yin/yang left behind for being underdeveloped, or just too far ahead of its time?

Or both - as first we had to work out reductionism, now we can go back to the project of holism.

I am just describing my own experience really. I learned the modern view of systems first. Then heard about Peirce. Then discovered that Anaximander, the first real philospher, had with surprising completeness got the whole essential systems story just about. And Aristotle - if you are reading him with a systems eye - was in fact struggling to marry the two perspectives. Anaximander's systems thinking and the later equally compelling worldview of atomism.

This is not of course how many people are taught Aristotle. History is told by the winners and so the whole of Ancient Greek philosophy 101 is about how these old fools lurched from one metaphysical extreme to another.

The early philosophers sought the fundamental substance (was it air, water, the apeiron?). Plato said no the fundamental was form (his version of substance, chora, barely gets mentioned). Heraclitus said all was flux (actually no, his view was more complex) and Parmenides bamboozled them by arguing there was only stasis, the impossibility of actual change. And instead of the illusory many, just the perfect one.

So it goes on. Every step of the metaphysical development hinged on discovering nature's dichotomies, but modern reductionism demands the story be taught as a series of monistic turns of thought.

So what I am saying is there is some stunningly well worked out systems theory in ancient greek philosophy. But no one really tells that tale.

The best academic account of Anaximander's philosophy is Anaximander and the Origins of Greek Cosmology by Charles Kahn - pretty dry of course.

 

[nismaratwork]

Or both - as first we had to work out reductionism, now we can go back to the project of holism.

I am just describing my own experience really. I learned the modern view of systems first. Then heard about Peirce. Then discovered that Anaximander, the first real philospher, had with surprising completeness got the whole essential systems story just about. And Aristotle - if you are reading him with a systems eye - was in fact struggling to marry the two perspectives. Anaximander's systems thinking and the later equally compelling worldview of atomism.

This is not of course how many people are taught Aristotle. History is told by the winners and so the whole of Ancient Greek philosophy 101 is about how these old fools lurched from one metaphysical extreme to another.

The early philosophers sought the fundamental substance (was it air, water, the apeiron?). Plato said no the fundamental was form (his version of substance, chora, barely gets mentioned). Heraclitus said all was flux (actually no, his view was more complex) and Parmenides bamboozled them by arguing there was only stasis, the impossibility of actual change. And instead of the illusory many, just the perfect one.

So it goes on. Every step of the metaphysical development hinged on discovering nature's dichotomies, but modern reductionism demands the story be taught as a series of monistic turns of thought.

So what I am saying is there is some stunningly well worked out systems theory in ancient greek philosophy. But no one really tells that tale.

The best academic account of Anaximander's philosophy is Anaximander and the Origins of Greek Cosmology by Charles Kahn - pretty dry of course.

Dry is fine as long as it's informative... I feel cheated by my education, which fell to precisely the traps and tropes you describe.

 

[apeiron]

Not air resistance, inertia.

Inertia? What is holding together the atoms of this bit of wall we are pushing about. There is a network of electrostatic bonds with an internal equilibrium to assert. We can only maintain a fiction of a localised reaction vector because the bit of wall does not fly apart into its atoms with the shove. If the wall did atomise, then we would have to chase after all the individual stories represented by the flying atoms (the now widely scattered "inertia").

You are of course familiar with the Machian mechanics debate as well?

Inertia just means mass, it doesn't mean not accelerating-- that's "inertial." Why those terms are used like that, I have no idea.

Inertia and mass do appear tied together. Though how this is the case is not completely straightened out (the Higgs mechanism is remarkably contextual wouldn't you say? )

And indeed Newton's laws are found wanting for just this reason.

You mean relativity fixed things by stepping back to more general symmetries - ones that could include spacetime as well its massive events.

So exactly as I have argued, Newton took his boundary conditions as static, eternal, uninvolved. This could not last. Einstein removed those specific constraints to model reality at a more general level - ones where those local values (the stress-energy tensor values) that make spacetime flat or empty have to be put back into determine the state of the geometry.

Newton's model was so constrained that it lacked flexibility. Einstein's model was less constrained and so constraints could be added back in as a choice. But it is still all the same trick - mechanics. You equilbrate away your global constraints to arrive at a model based on global symmetries. Then you add stuff back into this globally inanimate picture to animate it locally as required.

Talking about excellent reads, The Evolution of Physics by Einstein and Infeld is a great insight into how the mechanical view developed.

But there are thousands of must-read books. Sigh.

 

[Pythagorean]

Q_Goest

in addition to what was already said, you might want to consider what you know about a particular brand of differential equations in which the differentiation is no longer an integer value (so we can the 5.4th derivative instead of the 5th or 6th). If you have a network of "cells" and you couple them with such a term, it's not classical diffusion anymore, but it's still valid.

In this case, we no longer have nearest neighbor influences only. Each member of the system can now depend on whole system's global state rather than nearest neighbor. It has been shown recently that the diffusive process in highly turbulent systems is better described by this non-classical diffusive process (and we're still not talking about QM despite it being non-classical).

And of course, via my argument from before, these are open systems in reality, so you can introduce any kind of driving/forcing term you want to represent particular global effects. You're not going to be able to take all the different models describing different aspects and put them all together without conflicts and inconsistencies. They're models; they only work for what they were designed.

 

[Ken G]

I learned the modern view of systems first. Then heard about Peirce. Then discovered that Anaximander, the first real philospher, had with surprising completeness got the whole essential systems story just about. And Aristotle - if you are reading him with a systems eye - was in fact struggling to marry the two perspectives. Anaximander's systems thinking and the later equally compelling worldview of atomism.

I hadn't heard much about Anaximander, I'll have to find out more. I've long been interested in Parmenides and Zeno, and how they tried to invent a form of logic that could tell them things about reality, even if it told them the reality they recognize is an illusion. Amazingly, it kind of worked, as modern physics has found some remarkable synergy with the logical impossibility of change-- and the modern "quantum Zeno effect" must be some kind of record for the longest time between an idea and it's experimental confirmation, even if in an unanticipated way.

The early philosophers sought the fundamental substance (was it air, water, the apeiron?).

Which explains your handle...

So what I am saying is there is some stunningly well worked out systems theory in ancient greek philosophy. But no one really tells that tale.

I'm often struck by how many of the great questions they anticipated. It's almost impossible to find territory they didn't touch on somewhere. Ironically, they end up getting bashed for it-- so many questions, so few answers. People don't understand the most important thing philosophy does is map the terrain, you have to find your own destinations.

 

[Ken G]

Inertia? What is holding together the atoms of this bit of wall we are pushing about. There is a network of electrostatic bonds with an internal equilibrium to assert. We can only maintain a fiction of a localised reaction vector because the bit of wall does not fly apart into its atoms with the shove.

If the wall flies apart, the reaction vector might not be localizable into a single one, but there will still be reaction vectors. The forces that appear will depend on the larger context, but not the presence of reaction vectors-- in Newton's; scheme, every force always comes in pairs, whether the substance shatters or not. The internal forces of which you speak only affect the global context that determines how the forces play out, but not their coming in pairs-- the latter is purely reductionist, it's a sum of parts.

If the wall did atomise, then we would have to chase after all the individual stories represented by the flying atoms (the now widely scattered "inertia").

True, it would not be easy to make an accounting of all the action/reaction pairs there. But the Newtonian system says that they would be there all the same, and no matter how you group up the halves of the action/reaction pairs microscopically, the macroscopic result will always also be an action/reaction pair, because it is a simple sum.

You are of course familiar with the Machian mechanics debate as well?

Yes, the idea that mass there provides inertia here. But we're in the Newtonian system here-- explaining the source of inertia is not included!

Inertia and mass do appear tied together. Though how this is the case is not completely straightened out (the Higgs mechanism is remarkably contextual wouldn't you say? )

I'm not even sure the Higgs mechanism explains it.

So exactly as I have argued, Newton took his boundary conditions as static, eternal, uninvolved.

Yes, that's true, I think your point is valid that Einstein's view is more of a systems view, because the spacetime backdrop is itself embroiled in the action. In fact, that's not the end of it, because Einstein generates differential equations, so are open to the need for additional boundary conditions-- more systems. I was just saying that the law of action/reaction percolates up from the atomistic foundation of Newton's approach, it's bottom-up. Had Newton been completely right, there wouldn't be much room for a systems approach to any aspect of physical reality. Perhaps an Anaximander fan in Newton's own day could have been skeptical that reality could exhibit rich phenomena, like conscious choice, in such a sterile scheme, but that was Newton's scheme all the same.

 

[apeiron]

If the wall flies apart, the reaction vector might not be localizable into a single one, but there will still be reaction vectors.

I'm not sure whether you are agreeing or not. How do you account for a rocket ship for example? Do you try to sum up a bunch of tiny force vectors for the hot plume of combusting gas, or just go with the simpler single vector for the mass flow rate?

But the point I was making was about how the systems view is actually smuggled into mechanics.

Action~reaction is an example of how a global constraint (the presumption of energy conservation) has a downward causal effect(!) on the locales of a Newtonian system. That global symmetry entails the local ones. So any time something is seen to happen (an acceleration), there has to be a localised re-equilbration, a local conservation of energy. And this justifies the very simple approach of representing the situation as a pair of identical cancelling force vectors.

The same mechanical trick is repeated elsewhere.

In GR, of course, the law of conservation is not hardwired in as a global symmetry of the model. Instead, it has to be built in as a futher constraint - such as by specifying an inertial reference frame. So GR relaxes a global constraint to make the baseline model all flexxy, then allows you to put back in the constraints by hand to stiffen it up again and enforce a behaviour on a systems locales.

In QM, we have the problem of fluctuations - potential actions without a cancelling reaction. People start to think we might be able to turn the zero point energy into a perpetual motion machine! But fluctuations are tamed by the toy trick of virtual pair production. We say no, global conservation of energy still must rule. So what is actually happening down there (wink, wink) is that the vacuum is producing self-cancelling particle pairs.

The same with super-conductors. A real headache to model until BCS and the pretence that electrons joined hands to dash about as coupled bosonic pairs.

Mechanics is the sub-set of systems theory where the global constraints are treated as an equilbrium state that enforces also a local equilbrium. Once you have got the mechanics set up like this, a baseline view founded on a pervasive symmetry, then you can start modelling the propagation of change as symmetry breakings.

If the symmetry breakings are localised in some way, then you get a kind of quasi particle description of nature. If the symmetry breaking is global, well you get the big bang, the thermal model of time, etc.

(Talking of essential books again, Robert Laughlin's A Different Universe is a great polemic against the currently dominant reductionist mindset of physics).

 

[Ken G]

I'm not sure whether you are agreeing or not. How do you account for a rocket ship for example? Do you try to sum up a bunch of tiny force vectors for the hot plume of combusting gas, or just go with the simpler single vector for the mass flow rate?

You won't do the former, but you could-- the latter is the sum of the former parts. That is why you know the latter will be an action/reaction, because its pieces are. Newton's prescription is purely reductionist, it involves identifying the fundamental binary interactions, one by one, and summing them up, so any global conservation law stems from rules about those fundamental interactions. Later, physicists were able to see that the origin of these rules could also be viewed as global constraints (symmetries and conservation laws), but the reason the rules could be expected to apply to the fundamental interactions was the quintessentially reductionist principle that every interaction played out exactly the same whether its elements were part of a larger system, or if they were themselves the entire universe, individually subject to the global constraints. The universe in the head of a pin-- that is a fact about all of the force laws invoked in Newton's program.

Now, I think you are arguing that this is more of a bug than a feature, because the idea that every interaction in a complex system must be the same as it would have been had the elements been the whole universe misses out on how global constraints back-react on the elements. I think that's a valid point, but it is something quite missing from Newton's scheme-- it's not the reason behind the third law, all his laws are blind to it, and all would work without it. In a universe where nothing at all happened that was not understandable in a purely reductionist way, Newton's laws could describe it all (never mind relativity and quantum mechanics, those are detailed breakdowns of Newton's laws-- what you have in mind is a much more fundamental lacking element).

So I'm not disputing your point that conceptualizing all interactions as binary action/reactions at a distance misses what is going on at the systems level, I'm just saying that none of Newton's laws require a systems level to operate, they are perfectly self-consistent reductionist laws. I think that's what made them so seductive. The reason we have a systems level is not because we need it to get Newton's third law, it's that we need it to get the larger context of what richer type of behavior is possible than Newton's three laws. People who point to those three laws and say, "but I can get everything the system is doing just from those laws, oh and some appropriate boundary conditions and perhaps an externally applied time-varying field because I know that I'm going to need all that" are just ignoring how their fingerprints are all over the result, shoehorning the systems-level behaviors like a Greek playwright invokes the obligatory "deus ex machina" to make the end play out as desired. That's the part I agree with, the important part of the case you are making.

In GR, of course, the law of conservation is not hardwired in as a global symmetry of the model. Instead, it has to be built in as a futher constraint - such as by specifying an inertial reference frame. So GR relaxes a global constraint to make the baseline model all flexxy, then allows you to put back in the constraints by hand to stiffen it up again and enforce a behaviour on a systems locales.

Yes, it's that deus ex machina again.

Mechanics is the sub-set of systems theory where the global constraints are treated as an equilbrium state that enforces also a local equilbrium. Once you have got the mechanics set up like this, a baseline view founded on a pervasive symmetry, then you can start modelling the propagation of change as symmetry breakings.

The way I would frame that is, that's what mechanics looks like from the systems perspective. Mechanics can be set up from the reductionist perspective instead, which it normally is, but the advantage of seeing it from the systems perspective is that it immediately empowers you to relax the constraints of the model to encompass system-like behavior when you want to do that. Again it's the seductiveness of reductionism that shuns relaxing those constraints, almost like a person in deep water wishing to hold tight to the flotation device.

If the symmetry breakings are localised in some way, then you get a kind of quasi particle description of nature. If the symmetry breaking is global, well you get the big bang, the thermal model of time, etc.

Yes I think that's a useful insight, so I'll let this stand repeating:

(Talking of essential books again, Robert Laughlin's A Different Universe is a great polemic against the currently dominant reductionist mindset of physics).

 

[nismaratwork]

More books, and more debate... if I could somehow express clapping my hands together in girlish glee (a disturbing sight I assure you), I would.

 

[Q_Goest]

Hi Ferris,

If we want a mental event M1 to cause a physical event P2 and if we want the causal status of the mental to derive from the causal status of its physical realizer P1 (so that the theory doesn't fall in the substance dualist category) we are faced with over-determination (P2 could be realized by M1, as well as by P1 alone). If there are no greater causal powers that magically emerge at the higher level of M1 (if we want the theory to stay a materialistic one) then the causal powers of M1 are identical to the causal powers of P1, which means that P1 is the only realizer of P2, thus M1 becomes epiphenomenal. You can read more about this http://www.iep.utm.edu/mult-rea/#H4".

So, in the materialistic view you can either have mental causation identical with the physical causation or you can embrace epiphenomenalism and qualia. In both ways free will is impossible. If you want to find free will, you must seek it outside the materialistic domain.

Q_Goest,

In your post https://www.physicsforums.com/showpost.php?p=3179362&postcount=90" you say you don't believe in the phenomenal-physical correlation and basically you reject epiphenomenalism. And at first it doesn't looks logic, how can one make a knowledge claim about consciousness if it's epiphenomenal? But does the agent's association of the conscious experience of some event and its labeled state in the brain contradict in any way? The definition of the word "consciousness" in the brain state is not associated with the experience of it, but does this interfere the brain to be able to label certain physical state? Think about it, how will you explain the word "consciousness" to a little boy and what association does his brain make. For me epiphenomenalism implies that in exactly every millisecond your brain takes the optimal decision based on the available information. Even when you do something anti-evolutionary (take a lot of drugs, commit a suicide) it must be somehow justified in your brain calculations. Because if it's not, epiphenomenalism is wrong (remember you don't have taken the drug because YOU liked it, but because your BRAIN liked it).

I’m not too sure I really understand your point exactly, but I think you’d like to discuss the knowledge paradox a bit and I think your previous post to Ken is a perfect lead into that paradox. Note that I’m not presenting my opinions as much as I’m trying to maintain logical rigor here. I’m not suggesting that I reject or embrace epiphenomenalism. I’d like to point out one issue the epiphenomenalist argument must address but at this point there seems to be a fault in the logic.

Frank Jackson (Epiphenomenal Qualia) is a highly cited paper, being reference over 1000 times. He makes the argument that phenomenal properties such as qualia are not phenomena that can be described by describing the physical information on which the phenomena supervenes. I think this is a perfectly clear and legitamate argument. For example, we can describe physical information about any given physical phenomena such as how fast a guitar string vibrates, the tension in it, pressure waves created in the air, the vibrational interaction between the string and guitar, or how the bonds within the steel string change length as the string stretches and how the mass and inertia of the string causes it to move at a given frequency over time. Once we describe ALL the physical information about how the guitar works we’ve exhausted all there is to describe and there really is nothing left. How any physical system changes over time can be described by describing the physical information, and once that is done, there is nothing left to describe because we’ve fully described everything. But as Jackson points out, for what we call qualia or phenomenal consciousness, we haven’t described THOSE phenomena. A hypothetical neuroscientist might be able to describe everything there is to know about our nervous system, how our brains work, how neurons interact, how glia support neurons, now neurotransmitters work on a molecular level, etc… but we still haven’t described what the guitar sounds like, how a rose smells, or any other phenomenal property.

Let’s take the lead you provided from Kim regarding mental states (M) and physical states (P). For the causal closure of the physical, there are physical events P that determine other physical events. The mental events M are supervenient on the physical states but they don’t cause physical states. What causes physical states, assuming the causal closure of the physical, are other physical states. So the hypothetical neuroscientist that knows everything there is to know about our nervous system, can tell you what physical state P2 will follow physical state P1 (or what set of potential physical states will follow P1 if there is some random nature to them). Mental states that are described as phenomenal states are therefore epiphenomenal on the physical state. The mental state doesn’t cause the physical state, the physical states are caused by other physical states.

Epiphenomenal however, means that not only do these mental states not cause physical states, they also don’t influence them. They don’t have any way of influencing or causing a change in a physical state. If mental states were being “measured” by the physical state, they would no longer be epiphenomenal, they would suddenly become part of the causal chain that created the following physical state, so epiphenomenal in this regard means they really have no influence whatsoever over any physical state. So the paradox is, how do we know these states exist? The only reason given is that there is a 1 to 1 relationship between P and M, but that means we aren’t saying that we experience qualia because we actually experience that qualia. It says we are saying we experience something because of the physical states that cause us to utter those words.

Shoemaker, “Functionalism and Qualia” 1975:

To hold that it is logically possible (or, worse, nomologically possible) that a state lacking qualitative character should be functionally identical to a state having qualitative character is to make qualitative character irrelevant both to what we can take ourselves to know in knowing about the mental states of others, and also to what we can take ourselves to know in knowing about our own mental states.

Rosenberg, “A Place for Consciousness”:

Shoemaker is worried that, if functionalism is false (and certainly if physicalism is false), the relations between brain states and conscious states will be accidental in that the qualia involved in consciousness would make no contribution to determining our brain states. Because our brain states drive our behavior, including our knowledge claims, it seems that qualia would be irrelevant to what we could or could not claim to know.

...

Given that we are capable of making knowledge claims about consciousness, we need to understand how consciousness could be relevant to the production of those claims. To connect consciousness to the production of our claims about it, somewhere in our explanation of our knowledge we will need to appeal to the effects of consciousness on brain states. Now these brain states are solidly physical, and we are assuming the causal closure of the physical meaning that nothing nonphysical can make a causal difference. But if consciousness cannot affect brain states, it cannot play any part in producing our claims about it, and so it seems that we could not really know about consciousness. Yet we do know about it. Hence, Liberal Naturalism is caught in a paradox.

I am stating the intuitive problem. The Liberal Naturalist seems committed to conceding that consciousness makes no contribution to the fact that we make the claims about it that we do, and that is deeply troubling. Because any accuracy in our claims about it would seemingly be based on fortuitous coincidence, it seems impossible that we could know about it.

 

[Q_Goest]

Hi Pythagorean,

But think about this: let's say you have some giant system of N differential equations to describe the whole universe. You have every single interaction reduced to a handful of variables. Now all you need to do is put in your initial conditions for those variables.

What do you do? Your theory already accounts for everything in the universe, yet your theory doesn't account for how the initial conditions arose. Do you make the initial conditions a function of some part of the system? So now there was always this loop and never a beginning or end? I'm puzzled, personally, I have no idea what I'd do.

Anyway, I'm hoping this demonstrates that the science and the philosophy are completely different, just like models and reality. As another examples, we know that quantum mechanics underlies all classical observations, yet we naively model things in the old classical view. Why? Because it's effective, it's productive, it works. This is not the same way I approach the problem in a philosophical setting.

Not sure what the beginning of the universe has to do with. We don't have a theory of how the universe began, so let's not even consider it.

Regarding models and reality, it seems very confusing to me why you seem to feel that the philosophy of weak emergence is a "model" and not reality, but some sort of strong emergence or downward causation is not the model but is reality itself. Also, I can't help but wonder if your views of the dynamic systems approach are actually the same as those of the published work or not and if you really understand what the issue is. For example, do you think molecules in a fluid (say Benard cells for example) are causally affected not only by their local interactions with other molecules and with gravity but also causally affected by what occurs in distant parts of the fluid? What limits how quickly the molecules of a fluid at one point can affect molecules at a distant point? And what do you think is meant by the commonly used phrase, "the whole is greater than the sum of the parts"?

 

[Ken G]

Thanks for that clear description of what epiphenomenalism, and the "knowledge paradox", are all about. I don't think that Rosenberg's logic on the issue is solid. For example, when he says "But if consciousness cannot affect brain states, it cannot play any part in producing our claims about it, and so it seems that we could not really know about consciousness", there would seem to be a magical step in his argument: the step where he connects having knowledge about consciousness to having consciousness. I don't see those things as necessarily the same, so I see no paradox to the stance that a consciousness could emerge from a physical state, and knowledge about consciousness could also emerge from that same physical state. Indeed, that would seem to be an inevitable aspect of any flavor of physicalism. If both emerge from the same physical state, there is clearly no paradox in both having consciousness, and having knowledge about consciousness, without either affecting or altering the physical state at all.

But I don't buy it for other reasons. To me, physicalism, and functionalism, are both examples of putting the cart before the horse. We don't develop a physicalist, or a functionalist, viewpoint because we have any evidence that the universe really works that way, we do it simply because it succeeds in accomplishing the goals we have set out for the process. In a nutshell, if we establish physicalist goals, then physicalism is the path that leads there most economically. But whence comes the idea that this somehow means everything is physical? It's just bad logic to claim that follows, though one is certainly welcome to adopt it as a personal belief system, as with any religion.

However, a more natural stance, it seems to me, is that if one is interested in a physical question, like what are the neural correlates of some qualia, one should adopt a physicalist perspective, as one will not know where the blood is going by introspecting. But if one is interested in an epiphenomenal question, like what does pain feel like, one should adopt an epiphenomenal perspective, like introspection on the issue, because one will never know by watching blood flow what pain feels like.

As for which leads to which, again I see no evidence that a physical state leads to the qualia associated with it. Instead, it seems natural that both the physical state, and the epiphenomenal qualia, derive from something else, something we might consider to be what is Actually Happening There.

The physical correlates of the mental state are nothing but answers to a particular type of question about that state, and the idea that they are what "leads to" the mental state is an error in language, in my view. That's because what language is, above all, is connecting things to our experience. That's it, that's all language ever does-- it connects a phenomenon to our experience. What else can language do? Now, if our experience is always, at some level, a qualia, then the idea that physical states lead to the qualia has the situation exactly backward. Instead, when we speak of particles and potentials, we are using language, which connects to the qualia they trigger. We manipulate the qualia in terms of rules, or laws of physics, and generate outcomes, which are also qualia, which we then translate back to something we can compare to an experimental outcome, which is also a qualia. Somehow, in all that process, we are left with the idea that the qualia are derived from the physical states, but without the experience of qualia, the language we use to even talk about the physical states has no meaning at all. So it would seem that it is the qualia that lead to the physical states, in the sense that epiphenomenalism predates physicalism. The physical states do connect via a concept of causation, as Q_Goest described above, but as Hume so famously put it, we have no idea what causation is other than the observed tendency for one thing to follow another. In short, there is an algebra of qualia where one often follows another, and if we frame that in physicalist language, we can gain power over those causal connections whose origin is so mysterious to us.

In short, I would say the knowledge paradox has things backward-- the question is not, how can qualia matter if they never affect the physical states, the question is, how can what is physical matter if everything that we know about the physical is derived from our ability to experience qualia that we associate with a physical world? Instead of the qualia being a figment of our imagination, it is much easier to argue that the physical world we use the qualia to imagine is the thing we are imagining, albeit an imagination that follows rules outside our control. In short, the physical world is the thing we imagine when we want to imagine things that follow rules and are predictable, and what we call "our imagination" are the things we imagine when we don't impose that requirement.

 

[Ferris_bg]

Let's not derive from the main topic of the thread a lot with the knowledge paradox.

The only reason given is that there is a 1 to 1 relationship between P and M, but that means we aren’t saying that we experience qualia because we actually experience that qualia. It says we are saying we experience something because of the physical states that cause us to utter those words.

Ken G said it very well:

For example, when he says "But if consciousness cannot affect brain states, it cannot play any part in producing our claims about it, and so it seems that we could not really know about consciousness", there would seem to be a magical step in his argument: the step where he connects having knowledge about consciousness to having consciousness.

According to the epiphenomenalists the physical states that cause the agent to utter certain words do NOT put the agent's qualia in their "equations", but this does not interfere the physical states to be able to define the word "consciousness" (associating it with certain neural firing).

 

[Ken G]

According to the epiphenomenalists the physical states that cause the agent to utter certain words do NOT put the agent's qualia in their "equations", but this does not interfere the physical states to be able to define the word "consciousness" (associating it with certain neural firing).

Yes, it would seem that one is welcome to explore both consciousness and free will from the perspective of its neural correlates, but doing so will only help to answer certain types of questions. The physicalist has the neat solution of simply discounting all other questions as irrelevant, but by doing so, they forfeit the ability to answer a host of issues that are clearly relevant to the human condition, issues like morality, ethics, aesthetics, and what is a life worth living. All irrelevant issues to the physicalist, who can merely look at causation, and only through that looking-glass, darkly.

 

[Pythagorean]

Hi Pythagorean,

Not sure what the beginning of the universe has to do with. We don't have a theory of how the universe began, so let's not even consider it.

It has nothing to do with the beginning of the universe, just the general idea of a closed system vs. open system. If you can introduce a driving term to your model (such as a global electric field) even though you don't have a mechanistic model of how that driving term arose (you just know that you can measure it in the lab) then you've spoiled absolute reductionism (until you can provide a mechanistic description of the driving term, otherwise it's just some mathematical function you've added to make your system more descriptive.)

This goes for initial conditions too... we pick initial conditions as scientists, we don't have a theory for how they arose, so our system is always essentially open if it relies on initial conditions.

Regarding models and reality, it seems very confusing to me why you seem to feel that the philosophy of weak emergence is a "model" and not reality, but some sort of strong
emergence or downward causation is not the model but is reality itself.

Why do you keep saying this?

1) I've told you three times now (once in a previous thread) that I don't advocate strong emergence or strong downward causation. That's one of the many place where apeiron and I differ.

2) I've never made an attempt to say "this is the reality". Only that models are not reality, they approximate it, you can't prove a negative, etc, etc.

3) models that contain weak emergence are successful in describing reality. As I've said before, "strong" emergence seems like a wanting human interpretation.

please acknowledge that I've said this, since you've missed it three times before now.

Also, I can't help but wonder if your views of the dynamic systems approach are actually the same as those of the published work or not and if you really understand what the issue is.

I work as a dynamicist, for dynamicists, reading papers by other dynamicists (my most frequented journals are the journal of Chaos and Physics Review E).

For example, do you think molecules in a fluid (say Benard cells for example) are causally affected not only by their local interactions with other molecules and with gravity but also causally affected by what occurs in distant parts of the fluid? What limits how quickly the molecules of a fluid at one point can affect molecules at a distant point? And what do you think is meant by the commonly used phrase, "the whole is greater than the sum of the parts"?

Yes, they are affected by other parts of the fluid (in a turbulent regime, which is what most of nature is, even in biological systems). That is what the study of modern diffusion in turbulent transport is all about; fractalized diffusion terms (as opposed to classical diffusion). Anytime you have a fractalized derivative coupling your network components together, your behavior of one member no longer depends on just the nearest neighbors.

(by fractalized, I mean, as I said in a post above: the nth derivative is now the sth derivative. S is a real number, where n is integers, so you can have a 3.4th derivative). Using the fundamental theorem of calculus, derive the numerical form of a n=2 derivative, for instance. It will look something like V(m-1) + V(m+1) - 2(V(m)) (i.e. it will only depend on nearest neighbors, the m-1th cell and the m+1th cell). This is not the case if n is not an integer.

See the Hurst Exponent. When H = .5, you have classical diffusion.

We have one of the leading complexity experts working on this problem (fractal diffusion) at our university right now. He works on SOC and complexity in turbulent plasmas. My advisor (his partner) works on complexity in biological systems.

and this:

What limits how quickly the molecules of a fluid at one point can affect molecules at a distant point?

depends entirely on the system your modeling and what aspects you're modeling. You can't model everything at once in complex systems.