Maui said:The fundamental 'stuff' that everything is supposed to emerge from is still missing. For more than 100 years scientists have been failing to identify anything that resembles fundamental building blocks from which matter, time and space emerge(i reject strings and loops as idle speculation at this time and wavefunctions and Hilbert spaces as too ill-defined and ambigous mathematical tricks).
The topic of downward causation (and its opposite, causal reductionism) is an interesting one in its own right. But it would also be an interesting topic from the point of view of the sociology of knowledge. What I mean by this is, first, there are many ardent reductionists among philosophers and scientists, and I would state their position not in terms of “I have good grounds for this thesis,” but rather: “I can’t imagine how reductionism can fail to be true.” On the other hand, one can do a literature search in psychology and cognitive neuroscience and find hundreds of references to downward causation. Presumably these scientists would not use the term if they thought there was anything controversial about it.
Thanks apeiron. That's probably the one thing we'll always agree on! lolapeiron said:Yes, I am sure there is no way to change your mind here.
I don't see anything in those papers that would seriously suggest there are something like "top down constraints" that influence local causation. If conservation principals (conservation of mass, energy, momentum) are valid at every level, there is no room for downward causation, top-down constraints or any other uber, super premium level forces influencing local causation. It's all just weak emergence, and that's all we're entitled to (per Bedau).But anyway, boundary conditions would be another name for global constraints of course.
Immediately, when challenged, you think about the way those boundary conditions can be changed without creating a change. Which defeats the whole purpose. The person making the change is not factored into your model as a boundary condition. And you started with a system already at equilibrium with its boundary conditions and found a way to move them so as not to change anything. (Well, expand the boundary too fast and it would cool and the cells would fall apart - but your imagination has already found a way not to have that happen because I am sure your experimenter has skillful control and does the job so smoothly that the cells never get destabilised).
So FEA as a perspective may see no global constraints. Which is no problem for certain classes of modelling, a big problem if you are using it as the worldview that motivates your philosophical arguments here.
And as I said, a big problem even if you just want to model complex systems such as life and mind.
As asked, I provided examples of how top-down constraints such as selective attention have been shown to alter local neural receptive fields and other aspects of their behaviour. You have yet to explain how this fits with your FEA perspective where this kind of hierarchical causality does not appear to exist.
What is "it" you're referring to?Pythagorean said:Yes, I've seen the definitions, but my point was I guess, that I stand along side the people that think it's "magic". It seems rather mystical to me, which means either I don't understand it or it's bs. I chose to say I didn't understand it, I didn't mean that I didn't know the definition.
I accept there's "global behavior" as well, just as Benard cells for example exhibit a higher level behavior that "doesn't occur at smaller scales". But that doesn't mean there's a global orchestra conductor or any kind of functionally relavant top down constraints that alter what physically occurs at a lower level. Weakly emergent systems (as defined by Bedau for example) such as The Game of Life exhibit similar "global" behaviors. However, there's nothing but local, causal interactions that create those global behaviors. That's the philosophy behind FEA* and it's the philosophy behind the compartment models used in neuroscience today. Programs like NEURON and Genesis and the Blue Brain project demonstrate exactly the type of behavior we would expect given nothing but the weakly emergent rules set up by compartment models. There's no need for additional, higher level physical laws that somehow crowd out or take over the lower level laws. In fact, there are no such laws. When we talk about levels in nature, we're not talking about higher level laws, we're merely talking about the weakly emergent regularities that emerge from those lower level physical laws.I can definitely accept that there's global behavior that doesn't occur at smaller scales (a water molecule does not manifest a wave).
Q_Goest said:I accept there's "global behavior" as well, just as Benard cells for example exhibit a higher level behavior that "doesn't occur at smaller scales". But that doesn't mean there's a global orchestra conductor or any kind of functionally relavant top down constraints that alter what physically occurs at a lower level. Weakly emergent systems (as defined by Bedau for example) such as The Game of Life exhibit similar "global" behaviors. However, there's nothing but local, causal interactions that create those global behaviors.
Let's not get silly. Local causal interactions are well understood and modeled mathematically by engineers and scientists. Examples being the Navier Stokes equations, Hooke's law, etc... I really don't understand why that's a problem.Maui said:Local, causal interactions between what? To be certain that your worldview holds, you have to get to the bottom of it. Let's see what the biggest names in physics have come up with so far:
1. Non-local relativistic and deterministic wave structures(though 'relativistic' undermines the whole idea of cause and its effect)
2. Abstract fields
3. Your own fantasy
4. Our collective fantasy
5. Strings
6. Loops
7. Add anything you like
The only problem I see with this is that there is nothing 'functionally meaningful' that is gained by reducing classical scale phenomena (such as Benard cells or the interactions between neurons) to quantum scale interactions. By functionally meaningful, I mean that we choose to use classical descriptions because they account for the statistical agregate of all the particles (molecules) and there is nothing else in the configuration of those particles that can influence the system in a way that matters. No one is arguing that the classical description of the world is an exact one. The point is only that reducing a system of Benard cells or brain cells to the level of particle interactions doesn't gain us anything when we talk about the overall phenomena being studied.Ken G said:I think Maui's point is that even local causal interactions do not form a fully consistent ontology of the situation. So he/she might be saying that not only is there the issue that local causal interactions might be incomplete in regard to additional top-down mechanisms of the type apeiron is raising, it's even worse-- they are internally inconsistent in that they require we adopt an incomplete ontology even as far as local causal interactions go. This relates to the difference between computed trajectories and "true" trajectories, which raises the difference in the ontology of practical calculations versus what reality itself, even if we imagine it actually is deterministic, is doing. This seems a small issue, because the computed trajectories should mimic the "correct" ones in a statistical way, but there's always the chance that something underlying is going on that doesn't show up until you look for the right kinds of correlations.
Q_Goest said:The only problem I see with this is that there is nothing 'functionally meaningful' that is gained by reducing classical scale phenomena (such as Benard cells or the interactions between neurons) to quantum scale interactions. By functionally meaningful, I mean that we choose to use classical descriptions because they account for the statistical agregate of all the particles (molecules) and there is nothing else in the configuration of those particles that can influence the system in a way that matters. No one is arguing that the classical description of the world is an exact one. The point is only that reducing a system of Benard cells or brain cells to the level of particle interactions doesn't gain us anything when we talk about the overall phenomena being studied.
In fact, the "dynamic systems" approach would agree with this. That approach holds that there are classical scale interactions and "global constraints" that don't need to be reduced to the quantum scale. The emergent structures emerge BECAUSE of the classical scale interactions (ie: nonlinear ones, etc...). Not that I agree with that approach.
Q_Goest said:Let's not get silly. Local causal interactions are well understood and modeled mathematically by engineers and scientists. Examples being the Navier Stokes equations, Hooke's law, etc... I really don't understand why that's a problem.
Sorry, I really don't see anything deep here. Let's just say we disagree and leave it at that. :(Maui said:The problem is(and it's very deep) what those local causal interactions are between. The ontology of Newton is wrong and doesn't work. What does work, does not always favor causality and when it does, it involves more magic than weak and strong emergence...
Your model, whatever that may be, is just a model. It's neither reality, nor HOW reality is. It's awefully easy for anyone to shoot it down, as reality is much weirder than human imagination would accommodate. In philosophy, we are seeking(at least striving towards) truths and complete ontologies and naive models are always easiest to demolish. As one of the great thinkers once said - if your theory is not crazy enough, there is no hope for it.
Q_Goest said:Sorry, I really don't see anything deep here. Let's just say we disagree and leave it at that. :(
But this is just the issue-- how do we know a priori what is an "influence that matters"? You start by choosing what will matter to you, and this will then motivate the models you create, and when you will declare success. But when trying to model something like free will, when will you claim success? It seems very possible that when the only influences that matter are the ones that achieve gross bulk statistical behaviors, expressed within preconditioned degrees of freedom (to borrow from apeiron's language) and boundary conditions, you will not learn what kinds of special correlations might lead to quantitatively different outcomes, and will not know if you have succeeded because you might not even be trying to model the right things to get free will. I'm not saying you know you'll fail, I'm saying you can't know you'll succeed, and there might be reason, for some at least, to suspect you will fail.Q_Goest said:By functionally meaningful, I mean that we choose to use classical descriptions because they account for the statistical agregate of all the particles (molecules) and there is nothing else in the configuration of those particles that can influence the system in a way that matters.
For a second I had to check this wasn't coming from apeiron! I'm not saying we need to connect to the quantum scale, I view the quantum scale as simply an example of the kinds of unexpected correlations that emerge only when you know what to look for. For example, in a quantum erasure experiment, there is no hint in the raw data that any correlations exist there, they are embedded in the entanglements in ways that require clevel manipulation to extract. If someone happened to do a similar experiment prior to the days of quantum mechanics, they would have no idea whatsoever that they were missing anything using a classical mixed-state analysis-- and they might be tempted to use language about their assessment of their success that is similar to yours in the context of nonlinearly coupled systems.In fact, the "dynamic systems" approach would agree with this. That approach holds that there are classical scale interactions and "global constraints" that don't need to be reduced to the quantum scale. The emergent structures emerge BECAUSE of the classical scale interactions (ie: nonlinear ones, etc...). Not that I agree with that approach.
Q_Goest said:I don't see anything in those papers that would seriously suggest there are something like "top down constraints" that influence local causation.
Ken G said:I'm not saying we need to connect to the quantum scale, I view the quantum scale as simply an example of the kinds of unexpected correlations that emerge only when you know what to look for. For example, in a quantum erasure experiment...
Yes, I agree quantum mechanics forms an excellent example of this kind of surprising phenomenon, even if brains are not manifestly quantum mechanical. In quantum mechanics, everything happens that can happen, and it is only sorting of possibilities by the outer-scale environment that decides what actually does happen and brings the phenomena into the classical realm of black-and-white thinking. There remains no self-consistent ontology (without paying a radical price of subordinating the observed phenomena to how we conceptualize it) for describing how the microcausations that appear in the Schroedinger equation can create a single emergent classical reality. Decoherence is how we treat this event, but it involves deciding what we will care about, not an internally self-consistent treatment. I think that is also a relevant analogy to what FEA does-- first decides what we will care about, and then sees what harvest we reap from those choices.apeiron said:The classical world acts downwards to decohere the unformed potential of the quantum realm. There is nothing fixed at the QM scale, until it has been fixed.
I think the main source of disagreement is a choice of stance-- Q_Goest takes the position that he would prefer not leave a certain comfort zone, with demonstrable benefits, until he is absolutely certain it is required to do so, and what's more, he will tend to rig the meaning of "success" to increase its likelihood. We are taking the tack that it is better to jump into the unknown and murkier waters, and hope to discover something fundamentally new, than to avoid it simply because it is so murky.Q Goest mentions that this is the case...and then hurriedly says he doesn't want to go there. Indeed, it would be fatal to his insistence on causal supervenience.
Interestingly, even many-worlds is not really an atomistic construction, it is about as holistic as they come. It begins and ends with accepting that the concept of a "state vector for the universe" is a meaningful and coherent construct, because if you start with that, and trust the Schroedinger equation, then you will always have it (and its encumbent many conscious worlds emerging within), and if you do not start with it, then you will never have it appear. You get out exactly what you put in, and no experiment tells you if putting it in has done anything useful for you, so it's purely a desire to apply top-down imagery that motivates many worlds in the first place. Elsewhere, I've made the point that if you think about it, pure states don't propagate top-down, they propagate bottom-up: you get a pure state when you break a piece off from a larger system and force it to satisfy certain constraints, and there's really no other way that we ever encounter pure states in quantum mechanical analysis.(Unless he resorts to hidden variables, many worlds, the usual attempts to avoid the "weirdness" of downward acting constraints by leaping to far weirder views of nature that appear to preserve the mechanical principle that all that exists is bottom-up atomistic construction).
Q_Goest said:The only problem I see with this is that there is nothing 'functionally meaningful' that is gained by reducing classical scale phenomena (such as Benard cells or the interactions between neurons) to quantum scale interactions.
Ken G said:Interestingly, even many-worlds is not really an atomistic construction, it is about as holistic as they come. It begins and ends with accepting that the concept of a "state vector for the universe" is a meaningful and coherent construct, because if you start with that, and trust the Schroedinger equation, then you will always have it (and its encumbent many conscious worlds emerging within), and if you do not start with it, then you will never have it appear. You get out exactly what you put in, and no experiment tells you if putting it in has done anything useful for you, so it's purely a desire to apply top-down imagery that motivates many worlds in the first place. Elsewhere, I've made the point that if you think about it, pure states don't propagate top-down, they propagate bottom-up: you get a pure state when you break a piece off from a larger system and force it to satisfy certain constraints, and there's really no other way that we ever encounter pure states in quantum mechanical analysis.
Maui said:Just one example and i am leaving - the satellites that keep the GPS system working measure time differently, because it runs differently for observers in different referential frames. This is a fact backed up by hardcore science and thousands of expereiments. The implication of time 'flowing' differently is that your NOW has already happened(passed) in another frame of reference(e.g. that of the GPS clocks). It renders causality apparent(for some reason things seem(just seem) to have causes in the world of relativity). The other implication is that of free-will and free choice. It must also be just apparent. Add fields(the most consistent contemporary model we've build so far) and not just causality but everything observable is just excitations of a field(for some reason the excitations of the fields tend to conspire towards a seeming classical causality).
Q_Goest said:Hi Pythagorean,
What is "it" you're referring to?
I think you owe it to yourself to look at how neuronscience is treating the interactions of neurons at the level where computational modeling meets physical testing both disociated neurons and in vivo.
[...] there's nothing but local, causal interactions that create those global behaviors. That's the philosophy behind FEA* and it's the philosophy behind the compartment models used in neuroscience today.
And yet, it is really many-worlds which is holistic, and has a whole that is more than the sum of its parts. Many-worlds subordinates the physicist to the physics, and so invents a gossamer web of invisible coherences that act like glue between the islands of different worlds. The whole pure state is more than the sum of the "worlds", because the worlds lack these connections that the denizens of these worlds can never cross or even perceive. Some would call that holism on steroids, and prefer to subjugate the physics to the physicist. In that case, there is no need to restore anything that has been ruled out by the constraints of the "classical realm" (by which we both mean, observer effects). When physics is seen as the way a brain interacts with its environment, rather than the way an environment gives rise to a brain, there is no need for the concept of a unified state-- a state is merely whatever usefulness and consistency is left when all that is useless or inconsistent with constraints has been thrown out. So many-worlds is a glue, and Copenhagen is a sifter.apeiron said:Many worlds avoids the issue of this encounter because all destinies happen. So instead of a single global history, you just spawn endless micro-branches of history. Constraints break reality at a local level, atomising its history, rather than constraint being at the global level where it is more than the sum of its part and organising a single self-consistent history for reality.
An emerging idea is that physics, if not consciousness itself, involves that razor's edge between that which seems determined and that which seems random. That tennis match may echo your points about an interplay between bottom-up and top-down interactions.And this systems view also leaves room for the remaining quantum degrees of freedom that are the "weird" bit. When almost everything is being tidily regulated from the top-down, it can seem weird that the micro-scale is not actually under complete control but is still a little random.
nismaratwork said:Ummmm... it seems nobody else is going to break this to you, so I will... what the hell are you talking about? I understand your example, which is not one I'd use, nor does it have ANYTHING to do with causality, and free will in this context.
Ken G said: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.
Maui said:You should have probably first read my previous posts instead of yelling. Anyway, my latest post(that you quoted) was hypothetically taking into account a global scale view of reality(the so-called "God's eye view") and the role causality plays in view of relativity. Causality in the blockworld is not a fundamental feature of reality, instead - it's just an ordering of events in a causally-looking way when the observer is in a particular FOR in this Lorentz-invariant reality. As i stated earlier, this undermines the idea that things and events are really what they are because of causality(though they appear to be what they are because of causality). I stand by my words, causality very likely will not be a fundamental feature of a TOE, but just apparent/emergent(it doesn't really matter if everyone is already aware of this, but there is already a consensus on this among those who work on the foundational issues).
For causality to be fundamental, you'd need the universe of Isaac Newton. That universe is a mirage, however.
Maui said:If you've been reading Luboš Motl's blog, you've probably seen where the highest bets are being placed(which confirms what apeiron and i said earlier on emergence):
http://motls.blogspot.com/2004/10/emergent-space-and-emergent-time.html
Einstein was motivated by the search for something more fundamental, which is the highest goal of science. Why, then, do we feel the need to pretend that it was a search for something fundamental? Why can we not simply live in the truth?nismaratwork said:Hmmm... The very seach for what we like to imagine it to be is what set Einstein on his course to take a "Heuristic" view of light. It also hobbled him later in life...
... what to conclude from that?
Ken G said:Einstein was motivated by the search for something more fundamental, which is the highest goal of science. Why, then, do we feel the need to pretend that it was a search for something fundamental? Why can we not simply live in the truth?
Ken G said:But when trying to model something like free will, when will you claim success?
apeiron said:So arguing a modeling strategy works in some cases is not proving that it must work in all cases.
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.Pythagorean said:As we all point out though, the model is not the reality. The models are (successfully) predicting deterministic behavior of neural networks, it doesn't pretend to be the full story. It's a story about the electrical signal carried by the neurons. We don't consider, for example, . . . But they can, and especially if those behaviors are your interest.
I doubt that's any scientific field's prevailing philosophy, please provide clear evidence of this. I think this is your interpretation. Particularly that you put "nothing, but" as a qualifier. The prevailing philosophy is empiricism. To say there's nothing but local, causal interactions that create those global behaviors is a claim I've never seen. Whether it's true or not isn't important though. The interesting part of these complex systems is how the global behavior causes local interactions to occur, regardless of whether the global behavior originated from local interactions or not. We're not really interested in chicken and egg arguments in the lab; that's for philosophy forums.
The analysis may be approached from two different points of view. In one case, the forces acting on the members of the structure are considered as unknown quantities. In a statically indeterminate structure [think of this as your “top down constraint”], an infinite number of such force systems exist which will satisfy the equations of equilibrium. The correct force system is then selected by satisfying the conditions of compatible deformations in the members. ...
In the other approach, the displacements of the joints in the structure are considered as unknown quantities. An infinite number of systems of mutually compatible deformations in the members are possible; the correct pattern of displacements is the one for which the equations of equilibrium are satisfied.
The phrase “derivation by simulation” might seem to suggest that weak emergence applies only to what we normally think of as simulations, but this is a mistake. Weak emergence also applies directly to natural systems, whether or not anyone constructs a model or simulation of them. A derivation by simulation involves the temporal iteration of the spatial aggregation of local causal interactions among micro elements. That is, it involves the local causal processes by which micro interactions give rise to macro phenomena. The notion clearly applies to natural systems as well as computer models. So-called “agent-based” or “individual based” or “bottom up” simulations in complexity science have exactly this form. They explicitly represent micro micro interactions, with the aim of seeing what implicit macro phenomena are produced when the micro interactions are aggregated over space and iterated over time. My phrase “derivation by simulation” is a technical expression that refers to temporal iteration of the spatial aggregation of such local micro interactions.
:Dnismaratwork said:Um... as much as I'm dying to say, "[We] can't HANDLE the truth!", I wont... I did.
The search doesn't require adopting a belief system, it just involves doing science. The only required faith is the faith in the process, not the faith in the outcome. In fact, I would argue that faith in the outcome is what tends to close our minds to future advances, rather than the opposite. The same might be relevant to upwardly-causal approaches to free will.Anyway, why?... because that search has continually borne fruit such as QM, and Relativity, and perhaps elements of String Theory. The search itself tends to drive the field forward, but then the search becoming myopic is clearly crippling.
And indeed, elegance is certainly part of what we are seeking. We should look for elegance, and be happy when we find it, but not be seduced by it.Einstein certainly seemed to believe that there was an underlying elegance he could uncover, although it's true that he settled for "more" fundamental.
Right, and although it is natural to always try to find more fundamental unifications that correct inconsistencies, it is unnatural to expect that physics will ever be absent of inconsistencies. Never was, why do we imagine it ever will be?I'd add, the truth is ultimately inconsistant in the absence of a better means to join gravity and the forces described by QM.
Ken G said::D
The search doesn't require adopting a belief system, it just involves doing science. The only required faith is the faith in the process, not the faith in the outcome. In fact, I would argue that faith in the outcome is what tends to close our minds to future advances, rather than the opposite. The same might be relevant to upwardly-causal approaches to free will.
And indeed, elegance is certainly part of what we are seeking. We should look for elegance, and be happy when we find it, but not be seduced by it.
Ken G said:Right, and although it is natural to always try to find more fundamental unifications that correct inconsistencies, it is unnatural to expect that physics will ever be absent of inconsistencies. Never was, why do we imagine it ever will be?
Right, and we always think the ancient Greeks were so naive to put the Earth at the center! Still haven't learned that lesson.nismaratwork said:I don't know... I think people expect the universe to conform to an anthropic view.
Ken G said:Right, and we always think the ancient Greeks were so naive to put the Earth at the center! Still haven't learned that lesson.
Ken G said:Agreed. Now the question is, where does the scientific method lead into the study of free will, and do we need a few new tricks?