Exploring Free Will: Would Your Choice Be Different if You Could Choose Again?

[moving finger]

For both Vanesch and movingfinger or anyone else willing to discuss…

The sensation of free will, just like any qualia, is not objectively measurable in any way. There is nothing we can measure which says “here is free will” or “here is the color red” or pain, or any qualia.

I do not fully agree. Something which has only subjective properties is by definition not open to study through objective means. However, if we can identify the neural correlates of various states of consciousness, we may also be able to identify particular neurophysiological patterns which correspond to (correlate with) components of a “quale”. Such neural correlates would have objectively measurable properties.

I would like to propose that anything that is not measurable can have no measurable affect on that which is measurable.

interesting proposition, but as I argue above, it may be the case that we can identify some neural correlates of particular quales, in which case we have something we can measure and something which can be linked in a causal chain.

 

[moving finger]

Yes, I subscribe to this. Well, it is possible to subscribe to this, at least.

I understand what you are saying, I think. The question is: why are qualia, which cannot influence the physical state, determined by the physical state. A kind of violation of action-reaction.

We need to be careful with how we use the term qualia.
When you refer to a "quale", are you referring to an entire neurophysiological state (ONE aspect of which is that it produces some subjective properties - whereas the state itself also has objectively measurable properties)
Or, in using the term "quale", are you referring instead to ONLY the subjective properties of an entire neurophysiological state?
If the latter, then a quale is NOT an object per se, but is instead simply one of the properties of an object, and there is no reason (imho) why we should necessarily expect ALL properties of a physical object to be able to influence the physical world (ie some properties may be epiphenomenal).

 

[Rade]

...The sensation of free will, just like any qualia, is not objectively measurable in any way. There is nothing we can measure which says “here is free will”...

I do not agree. Free will is nothing more than volitional thinking. Thinking is not automatic or instictive or involuntary action--such as the digestion of food in stomach. To think is an act of choice, it is the act of integrating perceptions into concepts via a process called abstraction, and this process can be measured as an electro-chemical event by study of neutrons. So, not only can free will be measured, it also can be quantified. Here is one study that shows how activation of neurons as a free will response in relation to visual clues can be quantified:
http://findarticles.com/p/articles/mi_m1200/is_26_167/ai_n14839703

 

[Q_Goest]

vanesch said: The question is: why are qualia, which cannot influence the physical state, determined by the physical state. A kind of violation of action-reaction.

Yes. Interesting way of putting it. Regarding MWI, I’m not sure how that could be used to explain the correlation, but I enjoy your thinking.

movingfinger said: Something which has only subjective properties is by definition not open to study through objective means. However, if we can identify the neural correlates of various states of consciousness, we may also be able to identify particular neurophysiological patterns which correspond to (correlate with) components of a “quale”. Such neural correlates would have objectively measurable properties.

True. I fully agree that the ability to objectively measure a physical state, and that such a state might be having an experience is in fact possible. All views of consciousness accept this per the supervenience thesis including quantum mechanical, single cell neuron, EM field theories and all others. However, the computational paradigm has other limitations that some theories don’t have. The computational paradigm also assumes qualia emerge from classical mechanical interactions which are reducible to local, causal interactions at a macroscopic level. Thus, the computational model rules out any kind of strong emergence and any kind of downward causation unless 1) is shown to be false, which is impossible.

interesting proposition, but as I argue above, it may be the case that we can identify some neural correlates of particular quales, in which case we have something we can measure and something which can be linked in a causal chain.

Exactly. However, we must also realize that once we’ve allegedly identified these qualia by observing the classical level interactions of neurons, and of course, assuming we could exactly simulate those interactions using deterministic and random switches, we still have a problem.

The problem is that the switches in the computational model operate exactly as I’ve stated in 1) above:

1) For the case of a computational model, switches for example (or any classical interaction which can be duplicated in a control volume*) are influenced not by qualia, but by the voltage applied to them.

Once we realize that it is the voltages and not the qualia produced which is creating the behavior, we are stuck with qualia as being nothing but an epiphenomena. And once we’re stuck with qualia being an epiphenomena, we are stuck trying to explain why they should reliably correspond.

True story:
When I was just 4 or 5, I remember seeing a brightly polished, stainless steel pot on the stove. Mom was cooking dinner. The pot was shiney and reflective as a mirror. I remember bringing my face closer to it and watching my head suddenly get as large as a cantalope. But even when backing a few feet away, my head would get all skinny and almost disappear. I went back and forth several times, watching the reflection change shape just like one of those carnival mirrors. That was all well and good, but I was a strange child and decided to kiss the pot. The choice was all mine of course, and my mother came running as she heard the scream of her child in the kitchen and found him with scalded lips.

“Are you ever going to do that again,” my mother asked.

“Are you nuts? I’m never getting near that thing again!” I decided.

I felt pain (qualia – a bad experience).
I jumped back and screamed (behavior).
I felt a desire never to kiss another boiling hot pot again (qualia – free will).
I never kissed another pot again (behavior).

If we assume the computational model is correct, then we can:
a) do away with qualia – it isn’t needed to explain anything.
or
b) Suggest that qualia can be anything since the experience is not influencing behavior. (qualia does not need to reliably correspond)

From a) above, Dennett wishes to proceed with doing away with qualia, and he makes the implicit assumption that computationalism is true as he writes “Quining Qualia”.

From b) above, I’d like to also suggest qualia can be anything, and thus we have a problem as qualia DO seem to reliably correspond (quite literally). When pain occurs we exhibit behavior which resembles pain instead of something wonderful. If there is no correlation, I could quite literally love the pain of kissing boiling hot pots, but it would make no difference whatsoever in my behavior. I would still avoid kissing hot pots because the behavior is related to the action of voltages on switches and the qualia has no influence on that.

Similarly, I can experience the color turquoise when making a decision and wonder what the hell such a color has to do with not wanting to kiss pots any more. In the end however, when offered a chance to kiss another hot pot, the color turquoise could appear and I would say “Are you nuts?” My behavior is influenced only by voltage being applied to switches in the case of computationalism (or more appropriately, it is influenced only by classical signals). If it is only influenced by the voltage of an applied signal, there is either no reason to have an experience of free will when making a decision, or there is no reason to have any reliable correlation between the experience and the behavior.

We need to be careful with how we use the term qualia.
When you refer to a "quale", are you referring to an entire neurophysiological state (ONE aspect of which is that it produces some subjective properties - whereas the state itself also has objectively measurable properties)

Yes, exactly. We have a physical state (what you’re calling a neurophysiological state) and people readily assume this physical state produces some subjective properties which should correlate to the physical one. But I don’t see any valid argument that suggests these two states should correlate reliably given the computational model.

Hi Rade:

Free will is nothing more than volitional thinking. Thinking is not automatic or instictive or involuntary action--such as the digestion of food in stomach. To think is an act of choice, it is the act of integrating perceptions into concepts via a process called abstraction, and this process can be measured as an electro-chemical event by study of neutrons. So, not only can free will be measured, it also can be quantified.

I agree with anything you’ve said here. But note that what you’re talking about is the behavior of making a decision, not the sensation or experience of free will. When a switch changes position, or any group of switches change position, why should that correlate to the sensation or experience of making a decision. And if you say that experience we have with making a decision is X, then why can’t it be Y or Z? What difference would it make if we experienced Y or Z given the computational model? Please also read the above.

 

[moving finger]

1) For the case of a computational model, switches for example (or any classical interaction which can be duplicated in a control volume*) are influenced not by qualia, but by the voltage applied to them.

I think the problem with this statement, as I tried to explain above, is that it seems to implicitly assume that "qualia" are objects, rather than particular properties of objects.

If a "quale" is indeed an object which is influenced by other objects, then I agree we would expect that "quale" in turn to have some causal influence on other objects.

However, if what we call a "quale" is simply a particular (subjective) property of an object (which object has other objective properties apart from the "quale"), then imho there is no a priori reason why the "quale" (being simply a property rather than an object) should necessarily have any causal efficacy at all.

 

[Q_Goest]

Hi mf.

I think the problem with this statement, as I tried to explain above, is that it seems to implicitly assume that "qualia" are objects, rather than particular properties of objects.

Statement 1 is meant to say that qualia are a phenomena analogous to Brownian motion. The difference being that Brownian motion is an objective phenomena which reliably corresponds to the underlying mechanism. Brownian motion should correspond, even though it is an epiphenomena. It should correspond because there is an objectively measurable affect which molecules can have on a particle of dust.

So we could say that qualia are phenomena which have some kind of subjective property. For example, very high (or low) temperatures appear to produce the phenomena we call "pain". Similarly, the temperature of an object might produce any subjective experience whatsoever, such as an orgasm or the experience of turquoise, but they don't. Calling the phenomena of qualia a "property" isn't incorrect, but I want to be sure we agree on what this means.

However, if what we call a "quale" is simply a particular (subjective) property of an object (which object has other objective properties apart from the "quale"), then imho there is no a priori reason why the "quale" (being simply a property rather than an object) should necessarily have any causal efficacy at all.

So are you saying here that qualia should not have any causal efficacy (ie: is an epiphenomena)? If so, I'd agree that the computational model predicts this. And if this is the case, then the point we should be asking is why such phenomena should reliably correspond - or why they should exist at all, since an animal or human has no evolutionary advantage over a zombie animal or human.

 

[vanesch]

True story:
When I was just 4 or 5, I remember seeing a brightly polished, stainless steel pot on the stove. Mom was cooking dinner. The pot was shiney and reflective as a mirror. I remember bringing my face closer to it and watching my head suddenly get as large as a cantalope. But even when backing a few feet away, my head would get all skinny and almost disappear. I went back and forth several times, watching the reflection change shape just like one of those carnival mirrors. That was all well and good, but I was a strange child and decided to kiss the pot. The choice was all mine of course, and my mother came running as she heard the scream of her child in the kitchen and found him with scalded lips.

“Are you ever going to do that again,” my mother asked.

“Are you nuts? I’m never getting near that thing again!” I decided.

I felt pain (qualia – a bad experience).
I jumped back and screamed (behavior).
I felt a desire never to kiss another boiling hot pot again (qualia – free will).
I never kissed another pot again (behavior).

If we assume the computational model is correct, then we can:
a) do away with qualia – it isn’t needed to explain anything.
or
b) Suggest that qualia can be anything since the experience is not influencing behavior. (qualia does not need to reliably correspond)

To explain the externally observed actions (by your mother, say), indeed you do not need qualia. Probably the wirings of your neurons are such, that if the temperature of your lips rise, signals will be sent to your muscles to get them out of that situation, and a strong memorisation of the sensory information related to the situation will be stored in your brain. This will have as a consequence that next time similar sensory information is captured, that certain actions will not be performed anymore (like kissing a pot on a stove).

But all this can be done by a computer too. There's no need to invoque concepts such as "pain". Imagine you've designed a small robot piloted by your PC, with sensors such that when the temperature of a sensor reaches 50 degrees, it will now avoid next time a similar situation. You can, if you wish, call such a programmation "pain reaction". But do you really believe that your PC got hurt ?

So we are confronted to the difficulty that it is possible to explain all kinds of behaviour mechanistically, and that at no point we need "qualia" for that (unless we jokingly decide to call them that way). Your *mother* only assumed that it "really hurt" because she can imagine HERSELF doing the same thing, and, herself having subjective experiences, can imagine herself feeling pain, and then, by analogy, assume that her child has a similar experience.

You can say: "yes, but I TOLD her that it hurt". Now, imagine that you write a 5-line C-program that prints "aw! that hurt!" on the screen when you press the space bar. Even though your PC TELLS YOU that it hurts when you press the space bar, you won't assume that it "really feels pain", would you ?
Now, imagine that we've trained a kid in such a way that when we cut his hair, he screams that it hurts ! Do we believe him, or do we think he's kidding us ?

From a) above, Dennett wishes to proceed with doing away with qualia, and he makes the implicit assumption that computationalism is true as he writes “Quining Qualia”.

You can probably do away with qualia in order to explain *behaviour*. That still doesn't mean that we've explained a subjective experience !

From b) above, I’d like to also suggest qualia can be anything, and thus we have a problem as qualia DO seem to reliably correspond (quite literally). When pain occurs we exhibit behavior which resembles pain instead of something wonderful. If there is no correlation, I could quite literally love the pain of kissing boiling hot pots, but it would make no difference whatsoever in my behavior. I would still avoid kissing hot pots because the behavior is related to the action of voltages on switches and the qualia has no influence on that.

Exactly. So it seems to be a *property* of subjective experiences to be related to certain (which ?) physical phenomena, without them influencing it. Like it is a property of the sun to attract the earth. Call it "the law of the qualia" ...

Similarly, I can experience the color turquoise when making a decision and wonder what the hell such a color has to do with not wanting to kiss pots any more. In the end however, when offered a chance to kiss another hot pot, the color turquoise could appear and I would say “Are you nuts?” My behavior is influenced only by voltage being applied to switches in the case of computationalism (or more appropriately, it is influenced only by classical signals). If it is only influenced by the voltage of an applied signal, there is either no reason to have an experience of free will when making a decision, or there is no reason to have any reliable correlation between the experience and the behavior.

Except that this seems to be the case, so we could elevate it to some kind of law of subjective nature.

Yes, exactly. We have a physical state (what you’re calling a neurophysiological state) and people readily assume this physical state produces some subjective properties which should correlate to the physical one. But I don’t see any valid argument that suggests these two states should correlate reliably given the computational model.

Again, this being so then implies that there *is* such a connection, although we don't have a deeper underlying model for it that goes further than the postulate that "this is how subjective worlds behave". In the same way as we don't have a more fundamental explanation for why the laws of (objective) nature are the way they are, but it just happens to be so. Which is why we make this postulate.

 

[Q_Goest]

Hi vanesch,
I like your write up. Have a few reactions to it (in italic) and then some quotes I’d be interested in hearing your thoughts on.

To explain the externally observed actions (by your mother, say), indeed you do not need qualia. Probably the wirings of your neurons are such, that if the temperature of your lips rise, signals will be sent to your muscles to get them out of that situation, and a strong memorisation of the sensory information related to the situation will be stored in your brain. This will have as a consequence that next time similar sensory information is captured, that certain actions will not be performed anymore (like kissing a pot on a stove).

But all this can be done by a computer too. I agree that behavior can be mimicked by a computer There's no need to invoque concepts such as "pain". Imagine you've designed a small robot piloted by your PC, with sensors such that when the temperature of a sensor reaches 50 degrees, it will now avoid next time a similar situation. You can, if you wish, call such a programmation "pain reaction". But do you really believe that your PC got hurt ? no, don’t disagree with anything here.

So we are confronted to the difficulty that it is possible key phrase, “it is possible” to explain all kinds of behaviour mechanistically, and that at no point we need "qualia" for that (unless we jokingly decide to call them that way). Your *mother* only assumed that it "really hurt" because she can imagine HERSELF doing the same thing, and, herself having subjective experiences, can imagine herself feeling pain, and then, by analogy, assume that her child has a similar experience. agree. Computationalism does not attempt to give any value to qualia

You can say: "yes, but I TOLD her that it hurt". Now, imagine that you write a 5-line C-program that prints "aw! that hurt!" on the screen when you press the space bar. Even though your PC TELLS YOU that it hurts when you press the space bar, you won't assume that it "really feels pain", would you ? I don’t think a computer could ever experience anything.

Now, imagine that we've trained a kid in such a way that when we cut his hair, he screams that it hurts ! Do we believe him, or do we think he's kidding us ? You are appealing to my intuition about what my hair should be experiencing here. Maybe hair can experience something! If my hair’s experience is not subjectively measurable and is only an epiphenomena, then how am I supposed to determine if my hair is in pain or not?

I don’t think cutting hair can cause pain because it doesn’t have the proper physical substrate. Unfortunately, computationalists are forced to say that if the physical states of hair could be shown to correspond to physical states that cause pain, then cutting hair causes pain. They are forced to conclude this because of the concept of functionalism as originally provided by Putnam, though he now disavows this concept. If functionalism is proven false, then computationalism is false also. Arguments provided against functionalism often point out this problem that computationalism has. Putnam, Searle, Bishop and many others have pointed out this issue, and could show you why pain isn’t just caused by cutting hair, but cutting hair also causes every single experience imaginable. The arguments proposed are valid in my opinion. If computationalism is correct, cutting hair should create any and all experiences. Experiences don’t reliably correspond given computationalism. Putnam, Searle, and Bishop have been countered by Chalmers, Christley, Copeland and others - but that for another thread.

You can probably do away with qualia in order to explain *behaviour*. That still doesn't mean that we've explained a subjective experience ! agree. Note the difference between computationalism and other theories of consciousness which argue that qualia serves a purpose and is not just an epiphenomena.

Exactly. So it seems to be a *property* of subjective experiences to be related to certain (which ?) physical phenomena, without them influencing it. Like it is a property of the sun to attract the earth. Call it "the law of the qualia" ... Not sure why you’re referencing gravity. Gravity produces objectively measurable phenomena. Per computationalism, qualia doesn’t.

Except that this seems to be the case, so we could elevate it to some kind of law of subjective nature. I agree, “it seems to be the case”. “Why?” is a burning question.

Again, this being so then implies that there *is* such a connection, although we don't have a deeper underlying model for it that goes further than the postulate that "this is how subjective worlds behave". In the same way as we don't have a more fundamental explanation for why the laws of (objective) nature are the way they are, but it just happens to be so. Which is why we make this postulate. All other underlying laws of nature are measurable. I accept that Plank’s constant is what it is, simply because, but Plank’s constant is measurable and has a measurable affect. Brownian motion is measurable and is caused by an underlying mechanism. In each case we have reliable correlations. Saying “it just is the way it is” doesn’t say there is a reliable correlation if there is nothing measurable. It says we don’t know and we CAN’T EVER know. I don’t believe nature left us with this circumstance.

Take Chalmers for example as he talks about the possibility of strongly emergent phenomena:

As long as the existence of these phenomena is deducible in principle from a physical specification of the world (as in the case of cellular automaton), then no new fundamental laws or properties are needed: everything will still be a consequence of physics. So if we want to use emergence to draw conclusions about the structure of nature at the most fundamental level, it is not weak emergence but strong emergence that is relevant. [Referring to consciousness]
...
I think that even if consciousness is not deducible from physical facts, states of consciousness are still systematically correlated with physical states. [This is simply the restatement of the supervenience thesis]
...
In our world, it seems to be a matter of law that duplicating physical states will duplicate consciousness; but in other worlds with different laws, a system physically identical to me might have no consciousness at all [a physically identical system might be a p-zombie]. This suggests that the lawful connection between physical processes and consciousness is not itself derivable from the laws of physics but is instead a further basic law or laws of its own. The laws that express the connection between physical processes and consciousness are what we might call fundamental psychophysical laws.

Ref: Chalmers, “Strong and Weak Emergence”

Do you agree with this? If some phenomena (ex: consciousness) is not deducible in principal from the physical specification of the world then do you agree that this new phenomena must be strongly emergent or must rely on some other “fundamental psychophysical laws”? He points out cellular automatons as being weakly emergent because such things as CA's produce phenomena which are objectively measurable.

Note that Chalmers is a computationalist.

On the other hand, Henry Stapp is a particle physicist, so I think you can appreciate the fact he has a similar background and education to your own. Stapp however, feels that computationalism is faulty. He sees the same problems you and Chalmers talk about, the same problems we’re talking about here when he states:

One could imagine modifying classical mechanics by appending to it the concept of another kind of reality; a reality that would be thought like, in the sense of being an eventlike grasping of functional entities as wholes. In order to preserve the laws of classical mechanics this added reality could have no effect on the evolution of any physical system, and hence would not be (publicly) observable. Because this new kind of reality could have no physical consequences it could confer no evolutionary advantage, and hence would have, within the scientific framework, no reason to exist. This sort of addition to classical mechanics would convert it from a mechanics with a monistic ontology to a mechanics with a dualistic ontology. Yet this profound shift would have no roots at all in the classical mechanics onto which it is grafted: it would be a completely ad hoc move from a monistic mechanics to a dualistic one.

Ref: Stapp, “Why Classical Mechanics Cannot Naturally Accommodate Consciousness but Quantum Mechanics Can”

What do you think of Stapp’s perspective on this? Why should we “graft” on an ad hoc theory such as Chalmers’ “psychophysical laws” if in fact, there is no way to objectively measure the predictions made by these laws?

 

[ValenceE]

Hello to all,


I don’t think that the so-called free will, however you might describe it, is affected by the resetting of all parameters to a previous state.

Imo, so-called free will is a process that has no material self identity, coming in existence through our mental / emotional selves in the act of conscious decision making and therefore cannot be reset or affected by anything acting upon it.

In the proposed thought experiment, so-called free will would arise when the choice has funnelled down to coffee or tea, maybe being the first time it would be called upon during the morning, all other behaviour being mechanical-like.

However, imo, we are lacking information about the complete mental disposition of the subject and cannot derive the outcome as lightly as saying that a repeat choice = determinism, pure and simple.

What if, the night before, the person had just bought a new brand of tea that some friends raved about and that the choice was made, as he decided then, that he would have this tea for breakfast. It could be possible that, being awake for a little while, the subject didn’t quite connect with last night’s decision before the coffee or tea choice came up, and, then, thought to himself “oh yeah, let’s try this new tea “ thus choosing tea over coffee.

In this scenario, being put back in the morning’s setting would bring the same choice, not due to the determinism of the punctual situation, but from a previously made choice the night before, with a good time span between the two. Could we apply determinism here?

Or what if, the second time around, just before he would remember last night’s decision, his cat would spring by and take him out of the on going process, putting him in a different mental setting, leading him to do as he always does… take coffee. After a few sips he could then reconnect and think to himself “ shoot !, I wanted to try this new tea …”

Couldn’t this be two possible outcomes based on the same re-enactment ?


VE

 

[vanesch]

Take Chalmers for example as he talks about the possibility of strongly emergent phenomena:

Ref: Chalmers, “Strong and Weak Emergence”

Do you agree with this? If some phenomena (ex: consciousness) is not deducible in principal from the physical specification of the world then do you agree that this new phenomena must be strongly emergent or must rely on some other “fundamental psychophysical laws”? He points out cellular automatons as being weakly emergent because such things as CA's produce phenomena which are objectively measurable.

Yes, this is exactly what I also think. Subjective experiences are something which is strictly not derivable from "objective physical laws".

On the other hand, Henry Stapp is a particle physicist, so I think you can appreciate the fact he has a similar background and education to your own. Stapp however, feels that computationalism is faulty. He sees the same problems you and Chalmers talk about, the same problems we’re talking about here when he states:


Ref: Stapp, “Why Classical Mechanics Cannot Naturally Accommodate Consciousness but Quantum Mechanics Can”

Well, I also agree with Stapp! I don't see the contradiction, in fact. If I understand both well, then:
- Chalmers says that subjective experience is a "strongly emergent phenomenon" which is not deducible from objective physical laws ;
- Stapp says that in order to "model" subjective experience, we would need a dualist world, with an ad hoc postulated link from the "objective world" into the "subjective world".

I fail to see the difference: the second is a way to model the emergence of the first.

What do you think of Stapp’s perspective on this? Why should we “graft” on an ad hoc theory such as Chalmers’ “psychophysical laws” if in fact, there is no way to objectively measure the predictions made by these laws?

Just to have an ontological proposition which can resist all the objections to the materialist/behaviourist vision, without doing away with their advantages (no "mind over matter" difficulty in the observable physical laws), maybe ?

The reason why there are no objectively measurable predictions, is that this is not science, but meta-science (philosophy ?). So it is a good story which can reconcile our subjective experiences with objective physics. But we won't be able to verify it through scientific means. We should maybe accept that science and the scientific method has its limits too: it is limited to the objectively observable. We're talking here about something else, falling outside of this scope.

 

[moving finger]

Hi mf.

Statement 1 is meant to say that qualia are a phenomena analogous to Brownian motion. The difference being that Brownian motion is an objective phenomena which reliably corresponds to the underlying mechanism. Brownian motion should correspond, even though it is an epiphenomena. It should correspond because there is an objectively measurable affect which molecules can have on a particle of dust.

So we could say that qualia are phenomena which have some kind of subjective property. For example, very high (or low) temperatures appear to produce the phenomena we call "pain". Similarly, the temperature of an object might produce any subjective experience whatsoever, such as an orgasm or the experience of turquoise, but they don't. Calling the phenomena of qualia a "property" isn't incorrect, but I want to be sure we agree on what this means.

Good point. My choice of the word "property" was poor, and I agree that "phenomenon" is a better word to use.

However, I am still not sure that qualia are epiphenomenal.

So are you saying here that qualia should not have any causal efficacy (ie: is an epiphenomena)? If so, I'd agree that the computational model predicts this. And if this is the case, then the point we should be asking is why such phenomena should reliably correspond - or why they should exist at all, since an animal or human has no evolutionary advantage over a zombie animal or human.

I think I am saying that qualia perhaps do not have any causal efficacy (ie it is perhaps an empirically observed fact), rather than they should not have any causal efficacy (ie a provable fact).

Agreed, why "should" any epiphenomenon exist at all, if by definition it has no causal efficacy? There is no answer to this - imho "why do they exist" is not a relevant question in the case of epiphenomena. There is no "purpose" to an epiphenomenon, it simply exists. An epiphenomenon is simply an empirically (perhaps subjectively) observable phenomenon of the world; it perhaps has no "purpose", but having no "purpose" does not rule out existence. Not everything in the world needs to have a "purpose", thus asking the question "why does it exist?" is not necessarily a relevant question for everything in the world.

As for qualia - are they epiphenomenal or not? I'm not sure. As you say, if they are epiphenomenal then a conscious human would have no evolutionary advantage over a zombie. Why then are we conscious rather than zombies? One of only two reasons : EITHER qualia are epiphenomenal, but zombies are physically impossible (ie qualia, though epiphenomena, are not contingent, they are a physically (nomically) necessary part of the world); OR qualia are not epiphenomenal. At the moment, I cannot decide which.

 

[moving finger]

I don’t think that the so-called free will, however you might describe it, is affected by the resetting of all parameters to a previous state.

Imo, so-called free will is a process that has no material self identity, coming in existence through our mental / emotional selves in the act of conscious decision making and therefore cannot be reset or affected by anything acting upon it.

Cool. What you are proposing is that free will does not supervene (in the philosophical meaning of supervenience) on the physical - ie dualism. I have no problem with this, so long as you understand you are advocating dualism.

we are lacking information about the complete mental disposition of the subject and cannot derive the outcome as lightly as saying that a repeat choice = determinism, pure and simple.

The simple fact that "we lack information" is no obstacle to performing the thought experiment, which is an "in principle" experiment (not necessary to be achievable in practice).

What if, the night before, the person had just bought a new brand of tea that some friends raved about and that the choice was made, as he decided then, that he would have this tea for breakfast. It could be possible that, being awake for a little while, the subject didn’t quite connect with last night’s decision before the coffee or tea choice came up, and, then, thought to himself “oh yeah, let’s try this new tea “ thus choosing tea over coffee.

The only way this could happen (given the constraints of the thought experiment, and barring quantum uncertainty) is if the "didn't quite connect" part does NOT supervene on the physical world - which once again is dualism.

Or what if, the second time around, just before he would remember last night’s decision, his cat would spring by and take him out of the on going process, putting him in a different mental setting, leading him to do as he always does… take coffee. After a few sips he could then reconnect and think to himself “ shoot !, I wanted to try this new tea …”

The entire physical world is reset - remember - including the cat. If everything supervenes on the physical then the cat would not "spring by" the second time round if it did not also "spring by" the first time round.

There are no two ways about it. The ONLY way (apart from quantum indeterminism) to arrange for a different outcome the second time around is if some component (eg the "will") does NOT supervene on the physical (ie dualism).

MF

 

[Delphian]

Due to the sheer volume of responses to this thread, I haven't had the time to read them all yet. So many or all of what I say may only be reiterations, in which case forgive me for wasting your time.

Questions like this, and philosophy in general, are unanswerable unless every term used that shows the slightest ambiguity have their meanings predecided and agreed upon.
For example, even the phrasing of this situation leaves no generally agreeable answer.

The spiritualist might explain this by pointing out that if the will is indeed free then it is unrewindable. Since your experiment resets my will, it obviously assumes that it is not free.

Then see it as an alternate universe or reality. Nothing changes, it is a duplicate, they occur simultaneously, there is no 'rewinding,' and nothing is forced upon your will, leaving it "free." this was the OP's intent i think.

That brings us to the meaning of the term "free will." Free of what what exactly? If 'ultimate' freedom is implied, this means there is no intervening factors, such as taste, time, place, preference, etc. And the decision becomes a mental coin toss, where the subject has complete apathy whatever the outcome. By definition then, surely the decision has a 50% chance of having different outcomes. It would be as random as a literal coin toss. (Does this undo the entire free will idea? is this a paradox of "free" will then?)
If by free, you mean free of influence by a will that is not your own, then the subjects own personal bias' reigns. At the moment of choosing in one reality, in the other reality the exact same moment occurs, every decision you have ever made leading up to that moment remains the same, every single mitigating factor in the room with you and every room have you ever seen. your decision will not change in that very second.

If the experiment was to be altered however..

Take the time it takes for the subjects brain to make a snap decision as an 'instant.' a point in time of no measurable magnitude.

Then setting one of the realities any amount of time, however infinitely large or small, forwards or backwards could result in a different decision, as it is then NOT the same reality. Something has changed which lead you to be there earlier or later than you should have been. The conditions of the experiment have changed, it is not a fair test and as a result the decision COULD change, thought it might not either way.

 

[moving finger]

That brings us to the meaning of the term "free will." Free of what what exactly? If 'ultimate' freedom is implied, this means there is no intervening factors, such as taste, time, place, preference, etc. And the decision becomes a mental coin toss, where the subject has complete apathy whatever the outcome. By definition then, surely the decision has a 50% chance of having different outcomes. It would be as random as a literal coin toss. (Does this undo the entire free will idea? is this a paradox of "free" will then?)
If by free, you mean free of influence by a will that is not your own, then the subjects own personal bias' reigns. At the moment of choosing in one reality, in the other reality the exact same moment occurs, every decision you have ever made leading up to that moment remains the same, every single mitigating factor in the room with you and every room have you ever seen. your decision will not change in that very second.

I think that a believer in (libertarian-style) free will would say that, even if one were to carry out the experiment and reset the conditions back to exactly how they were, the "will" would still be free to choose either tea or coffee - the outcome would not be predictable by another observer - in other words the "will" does not supervene on the physical. This leads to dualism.

If the experiment was to be altered however..

Take the time it takes for the subjects brain to make a snap decision as an 'instant.' a point in time of no measurable magnitude.

Then setting one of the realities any amount of time, however infinitely large or small, forwards or backwards could result in a different decision, as it is then NOT the same reality. Something has changed which lead you to be there sooner or earlier than you should have been. The conditions of the experiment have changed, it is not a fair test and as a result the decision COULD change, thought it might not either way.

Unfortunately this does not provide for any escape from the problem, unless one once again advocates either dualism or quantum uncertainty. The fact that we are dealing with a thought experiment means that we need only be able to reset the scene "in principle", and not in practice. An infinitesimal time period would be impossible to reset in practice, but not in principle.

There is no escaping the conclusion that the "will" either supervenes on the physical (in which case resetting the physical also resets the "will"), or it does not (in which case the will can operate independently of the physical, but at the cost of opening the door to dualism).

 

[Delphian]

There is no escaping the conclusion that the "will" either supervenes on the physical (in which case resetting the physical also resets the "will"), or it does not (in which case the will can operate independently of the physical, but at the cost of opening the door to dualism).

Then your question has no answer.
Unless you would like to choose one of these two as a given state in the experiment.

 

[ValenceE]

moving finger...

Cool. What you are proposing is that free will does not supervene (in the philosophical meaning of supervenience) on the physical - ie dualism. I have no problem with this, so long as you understand you are advocating dualism.

On the contrary, coming in existence ONLY during the mental thought process of making a conscious choice, 'free will' is intimately tied to the physical. No dualism here.

The only way this could happen (given the constraints of the thought experiment, and barring quantum uncertainty) is if the "didn't quite connect" part does NOT supervene on the physical world - which once again is dualism.

Why ?... why is NOT supervening the only way this could happen ?

There are no two ways about it. The ONLY way (apart from quantum indeterminism) to arrange for a different outcome the second time around is if some component (eg the "will") does NOT supervene on the physical (ie dualism).

Same here... why ? How can we say without a doubt that the thought processes including 'free will' couldn't both supervene and, WHILE supervening, possesses their own realm ?




VE

 

[moving finger]

why ? How can we say without a doubt that the thought processes including 'free will' couldn't both supervene and, WHILE supervening, possesses their own realm ?

Simple.

If A supervenes on B (in the philosophical sense of supervenience, see Wikipedia) then any change in A entails a change in B (definition of supervenience).

Thus, if we reset B to be the same as it was at a previous time, it follows that A will also be reset.

Let B = the physical world, A = human "will". Now in absence of (quantum) indeterminism, the physical world operates deterministically, hence resetting B will result in replay of the physical. If human "will" supervenes on the physical, it follows that replay of the physical will result in replay of human "will".

The only way to avoid the above conclusion is to introduce (quantum) indeterminism (but that alone doesn't lead to free will), or to claim that human "will" does not supervene on the physical.

 

[mtsweetpow]

Personally, I like to mix up my rational for making decisions. I drink a fair amount of coffee and tea and to the best of my self conscious being i cannot describe why I would choose one or the other.

Avoiding the desire to equate an answer to the question, I propose that free will is not inherant or oppositly, unachievable.

Rather, it is a phenomina that ebbs and flows based on pabst, present and predicted future conditioning of the brain.

While thinking to yourself, is each successive topic or thought dependent on those in the pabst?

I challange that free will can be learned, that it is a product of our consiousness.

It becomes apparent when giving credit to the creators of improvisational music, dance, mechanical design... that the human mind has an ability when prompted correctly to create new logic, new creation, new perception drastically different than the predicted or known outcome.

My humbly proposed (cant wait to hear responses!) answer is that a person more aware of and practiced in the ways of the minds conscious abilities will be more apt to free will and thus more likely to choose tea in the event of a time shift and deletion of elapsed experiences.

 

[mtsweetpow]

Personally, I like to mix up my rational for making decisions. I drink a fair amount of coffee and tea and to the best of my self conscious being i cannot describe why I would choose one or the other.

Avoiding the desire to equate an answer to the question, I propose that free will is not inherant or oppositly, unachievable.

Rather, it is a phenomina that ebbs and flows based on pabst, present and predicted future conditioning of the brain.

While thinking to yourself, is each successive topic or thought dependent on those in the pabst?

I challange that free will can be learned, that it is a product of our consiousness.

It becomes apparent when giving credit to the creators of improvisational music, dance, mechanical design... that the human mind has an ability when prompted correctly to create new logic, new creation, new perception drastically different than the predicted or known outcome.

My humbly proposed (cant wait to hear responses!) answer is that a person more aware of and practiced in the ways of the minds conscious abilities will be more apt to free will and thus more likely to choose tea in the event of a time shift and deletion of elapsed experiences.