What Price for Genuine Free Will?

[moving finger]

I agree with you, Einstein, Bohr & Schroedinger - i.e. QM is good for predicting experimental results (when the math can be solved) but what, if anything, it is stating about "reality" is very likely beyond human comprehension (too strange for us who have basically experienced only a classical world).

Perhaps we need to “let go” of the classical world.

Humans also have feeling, intuitions, emotions, hopes, etc. which are essentially impossible to define. My GFW & MG fall into this group.

I have a deep suspicion of anything that is “impossible to define” – this rings warning bells in my ears (suggesting that there is something amiss……)

Try study of attachment "free will - out of Africa." Ask questions about any parts of it you don't understand or disagree with.

OK, I will. Thanks.

I will accept your definitions, provided you allow the possibility that one can refine the definition of "indeterminism" to contain two "mirror parts" - one called "random decisions" and the other called "agent decisions."

Hmmmm. Are these “agent decisions” deterministic or indeterministic? I think you see the problem……

But it does show another way to frame the same old question is: Do genuine agents exist?

What is a “genuine agent”? Is such an agent operating deterministically, or indeterministically, or “something else”? If something else, then what?

Furthermore, I find it hard to believe that there are any significant QM processes going on in the human brain. (Some people hold hope that QM processes occur in "microtubes" / very fine "hairs" / that are attached to cells - but I am not interested in any "chance free will" made possible by QM "observations" forcing mixed state wavefunctions into Eigen states.

I agree with you. Your references are (I believe) to Roger Penrose, who is a great mathematician but (forgive me) a poor scientist and poor philosopher.

I suspect the human brain is a deterministic computer, even if QM does not have hidden variables etc.

I agree

Thus, from this and prior paragraph, if GFW is to exist, it can only exist in a non material form - either a "soul", which I don't want to turn to for my desired GFW, or in a simulation of the real world - the only world "we" actually experience and "live" in. - See attachment.

Hmmmmmm.

I agree one can separate software from hardware, but I do not agree that this is a good analogy for the concept of “self”. Let us imagine a “gedanken” experiment – Einstein’s brain. Let us imagine that some incredibly advanced alien race had managed to analyse the way that Einstein’s brain worked to the finest detail, such that they could reproduce his brain, in it’s entirety, on one of their computers as a program (in hardware plus software). ...

I anticipated this argument more than 10 years ago in paper about visual perception that had a brief philosophical speculative termination (My ideas about the possibility of GFW fell out unexpectedly from my revision of perceptual theory presented in this paper. - Ref 1 of the attachment to my first post here.)

Basically I claim that in any complex neural system, the physical connections are important but more important for thoughts are the ionic and neurotransmitter fluxes. These dynamic variables can not be recovered from a dead brain.

OK, I agree. After death I agree that many of the important synaptic junctions will change. But in my “gedanken experiment” I did not specify whether the aliens analysed Eintein’s “live brain” or his “dead brain”. It is after all a gedanken experiment.

I went so far as to postulate a "biological uncertainty principle" which basically states that the more carefully one measures the dynamic activity of a complex living brain, the more this activity is disturbed by the measurement process.

Agreed this would be a limitation in practice, but I suggest not a limitation in principle?

Thus I do not believe your "thought experiment" is possible even with great advance in technology. - I believe that my "biological uncertainty principle" is true for complex brains, but admit it does not have the mathematical proof that the physical one does.

OK, I agree you may be right here. Thus in practice it may be impossible to make a “copy” of a brain. But in principle?

MF

 

[Billy T]

... Are these “agent decisions” deterministic or indeterministic? ...What is a “genuine agent”? Is such an agent operating deterministically, or indeterministically, or “something else”? If something else, then what?

By agreeing to accept your definitions of deterministic and indeterministic, I gave up the "Middle Ground" concept, or accepted that there is zero space between these two domains for MG.

Thus the "agent" is indeterministic and non-random. That is, the agent exhibits "purposeful behavior" which can not be completely predicted, because it is not deterministic. (I am not confusing predictably with indeterminate, as some do.) I am, in essence, (if I am not talking nonsense - a real possibility), trying to claim that there is some currently unknown form of logic, which can be implementable in a deterministic computation facility (complex brains only at the present level of technology) that is not deterministic and yet not entirely random.

"Fuzzy logic" is close to the idea, but not actually what I want as it is entirely repetitive if simulaneously implemented in two or more identical computers. (That is, by looking only at one computer's results, you can predict exactly what the other is producing - a form of predictability, but neither's future acts can be predicted.)

That is, I don't think all human agents fail to be identical only because the "agent logic" is running in different brains and subject to different environments. If it were possible to have cell-by-cell identical development from the moment of conception in the womb and always thereafter plus completely identical environments, I suspect that you would (at least) lean to the idea that these "more than genetically identical" twins would have the same personality, but I would not. I would think that the "agent programs" in each would make different choices, despite these identical circumstances. This is another way for me to say that "agent programs" are not just some current form of "fuzzy logic." Identical currently-known fuzzy-logic programs running simultaneously in identical brains and subjected to identical inputs, would have the same personality.

That is, the "agent logic" I postulate exists is unique, but how this is achieved may well be a random selection. Probably the general structure is universal in humans, much like N. Chompski's genetic human language facility is.

...Agreed this would be a limitation in practice, but I suggest not a limitation in principle?...OK, I agree you may be right here. Thus in practice it may be impossible to make a “copy” of a brain. But in principle? MF

Neither of us can be certain if my "biological uncertainty principle" for complex brains is "true in principle," but I would argue for this along the following lines: There is a good possibility that the human brain is a chaotic system. By that I mean that one neurotransmitter molecule attaching to the cell wall of a neuron can tip the balance and cause a "action potential" to travel down it axion. The neurotransmitters released by this impulse in several hundred other synaptic junctions, then tips the balance in some of them etc. The brain is a fantastic amplifier of small events.

For example; a single Brownian motion controlled neurotransmitter event in Einstein's brain may have made him either (1) first speculate that the velocity of light is a constant. - When he did. If he had lived, See (2), he probably would have done so soon later. or (2) alternatively make him decide he needed a break and some exercise, which could have caused him to go for a bicycle ride. (That unfortunately for physic, gets him killed by a drunk man driving his Packard car up onto the side walk.)

Almost everything important in my life has seemed to have been the result of chance, beginning with my "choice" of parents and birth country. Firestone accidently dropped some latex with sulphur in it on a hot stove. But as he, or someone else, observed: "Chance favors a prepared mind."

 

[moving finger]

Thus the "agent" is indeterministic and non-random. That is, the agent exhibits "purposeful behavior" which can not be completely predicted, because it is not deterministic. (I am not confusing predictably with indeterminate, as some do.) I am, in essence, (if I am not talking nonsense - a real possibility), trying to claim that there is some currently unknown form of logic, which can be implementable in a deterministic computation facility (complex brains only at the present level of technology) that is not deterministic and yet not entirely random.

Beats me then. I have no idea what this might be. (I don’t want to sound flippant, but there is a possibility that what you are looking for might be akin to some miraculous magic-pill which cures baldness – sounds nice, but doesn’t actually exist)

"Fuzzy logic" is close to the idea, but not actually what I want as it is entirely repetitive if simulaneously implemented in two or more identical computers. (That is, by looking only at one computer's results, you can predict exactly what the other is producing - a form of predictability, but neither's future acts can be predicted.)

Agreed – except the future acts of one could be predicted by the other?

That is, I don't think all human agents fail to be identical only because the "agent logic" is running in different brains and subject to different environments.

Hmmmm. I do.

If it were possible to have cell-by-cell identical development from the moment of conception in the womb and always thereafter plus completely identical environments, I suspect that you would (at least) lean to the idea that these "more than genetically identical" twins would have the same personality, but I would not.

Although one might guarantee identical “nature” (ie genes), in practice it would be impossible to guarantee the “nurture” was identical.

I would think that the "agent programs" in each would make different choices, despite these identical circumstances.

OK. I do not. I think my mechanism is fairly clear. How does your mechanism work?

Neither of us can be certain if my "biological uncertainty principle" for complex brains is "true in principle," but I would argue for this along the following lines: There is a good possibility that the human brain is a chaotic system.

Agreed. But chaotic systems are still deterministic.

By that I mean that one neurotransmitter molecule attaching to the cell wall of a neuron can tip the balance and cause a "action potential" to travel down it axion. The neurotransmitters released by this impulse in several hundred other synaptic junctions, then tips the balance in some of them etc. The brain is a fantastic amplifier of small events.

Agreed. But still deterministic.

For example; a single Brownian motion controlled neurotransmitter event in Einstein's brain may have made him either (1) first speculate that the velocity of light is a constant. - When he did. If he had lived, See (2), he probably would have done so soon later. or (2) alternatively make him decide he needed a break and some exercise, which could have caused him to go for a bicycle ride. (That unfortunately for physic, gets him killed by a drunk man driving his Packard car up onto the side walk.)

Agreed. But still deterministic.

Almost everything important in my life has seemed to have been the result of chance, beginning with my "choice" of parents and birth country.

Agreed. But what you call chance may simply be your epistemic horizon (ie your lack of knowledge). It could all have been deterministic.

I grant you that the world may not be deterministic, but I still do not see how introducing indeterminism endows free will.

MF

 

[Billy T]

We are getting so agreeable the it is becoming predictable and perhaps uninteresting. You made 10 comments in your last reply, most starting with: "Agreed..." and the few that did not were ideas I agree with. E.G. It also "beats me" how GFW could work. Etc. That is, in 9 of your 10 we have the same view. Even in the other one, now quoted:

...OK. I do not. I think my mechanism is fairly clear. How does your mechanism work?

we don't actually disagree - I just have a different less well defended opinion. I readily admit (and have done so in several prior posts) that your position is "air tight," secure etc. and that my efforts to support the possibility of GFW are failures by comparison - only hopes expressed.

The reason I have not yet joined you in "your camp," as I have oft put it, is that neither you nor I have been able to firmly crush this hope (wish I could to end the uncertainty) - All either of us can muster is to say that because it is not well defined, it is too slippery to squash. If money were involved, mine would be bet on your ideas, not mine, but I do feel an attraction to my hope and since I lack the knowledge of what may exist in some logical structures and I do know of at least a self reflexive logic loop that I don't know how to analyse logically ("This sentence is false.") I can not abandon this ill defined hope.

I have twice in the past offered to stop our discussion and do so again. If we do, perhaps I can spend some time reading posts the the philosophy subgroup concerned about logic, epistemology, etc. and subsequently better formulate how the non-deterministic, yet non-random, "purposeful" logic an agent with my GFW must have, might work. Again I say I have enjoyed our exchanges. Your clear logic and knowledge have helped me to understand better my hope for GFW.

There is one lose end in our exchanges concerning your question about entropy. As I recall, you were some what at a loss as to why it does not decrease in closed system. (I think it can, just that it is very rare because of the way humans define the states or categories.) To illustrate this, I described a box with 100 white and 100 black marbles and used it to show that this law of physics (thermodynamics) has a great deal to due with the way we humans collect equally probable individual configurations into groups. (I also illustrated it with my only bet in Brazil's popular lottery - I bet on the numbers 1,2,3,4,5 & 6 (six are required from the 60 available) in a failed effort to get my wife to refrain from making weekly bets in this less than 50% payout game.)

You never commented on my idea that "entropy increase" is mainly, if not entirely, to due to the way humans assemble/define categories. - Only two of the equally probable outcomes have all marbles of one color in the same layer, (the "vertically divided" category) but millions of arrangements are in the "well mixed" category, so of course we are surprized if after shaking the closed box, the vertically divided category is observed when the box lid is lifted, but not surprized when one equally probable member of the "well mixed" categories is observed. If box is shaken and then observed enough times, there will be steps in the long chain of observations where entropy did decrease. (I know that the entropy of the system including the shaking mechanism has increased, but any believer in determinism and the ergotic theorem and chaos will probably understand what I am trying to state. - How we define our categories has a lot to do with the increase of entropy in simple systems like my box of marbles or lottery results.)

 

[moving finger]

I have twice in the past offered to stop our discussion and do so again.

OK

Your clear logic and knowledge have helped me to understand better my hope for GFW.

Thank you, I’m pleased.

There is one lose end in our exchanges concerning your question about entropy. As I recall, you were some what at a loss as to why it does not decrease in closed system. (I think it can, just that it is very rare because of the way humans define the states or categories.)

Agreed. If I said that it does not then I apologise – that is an error. The 2nd law is obviously a purely statistical law hence it does not say that entropy cannot decrease.

To illustrate this, I described a box with 100 white and 100 black marbles and used it to show that this law of physics (thermodynamics) has a great deal to due with the way we humans collect equally probable individual configurations into groups. (I also illustrated it with my only bet in Brazil's popular lottery - I bet on the numbers 1,2,3,4,5 & 6 (six are required from the 60 available) in a failed effort to get my wife to refrain from making weekly bets in this less than 50% payout game.)

You never commented on my idea that "entropy increase" is mainly, if not entirely, to due to the way humans assemble/define categories. - Only two of the equally probable outcomes have all marbles of one color in the same layer, (the "vertically divided" category) but millions of arrangements are in the "well mixed" category, so of course we are surprized if after shaking the closed box, the vertically divided category is observed when the box lid is lifted, but not surprized when one equally probable member of the "well mixed" categories is observed.

Agreed. Though I believe there are some (however) who would claim that this kind of “information entropy” is not in fact a proper measure of entropy (for precisely the reason that it depends on how one defines information).

If box is shaken and then observed enough times, there will be steps in the long chain of observations where entropy did decrease. (I know that the entropy of the system including the shaking mechanism has increased, but any believer in determinism and the ergotic theorem and chaos will probably understand what I am trying to state. - How we define our categories has a lot to do with the increase of entropy in simple systems like my box of marbles or lottery results.)

Poincare’s return. Boltzmann (I believe) had suggested that the low entropy past of our world may have been caused by such a statistical variation. However as I noted above, I believe some may disagree that statistical information measures are true measures of entropy.

MF