Aside from the trivial non-coercion meaning and the randomness that harms any kind of will, the "definition" eludes us since it never works out, so far, but the Holy Grail of the crux of it is to find a way above and beyond the automated brain will being true to itself that lets there be some higher agency that is somehow 'free' and 'independent' of the brain will or able to will the brain will, but, again, we not being able to well define this 'free' idea, much less to go on to show it. — PoeticUniverse

Of course, it's entirely impossible to identify something "free" in the physical world, using the scientific method. At best the scientific method can establish randomness. Freedom is definitely not physical.

You know...I always thought that, too. But then I came across this “degrees of freedom” for showing coordinate dimensions in a phase space, and I got to wondering how freedom was meant to apply there. I don’t consider freedom to be physical either, but apparently, somebody figured degrees of it, are.

The two AIs DO have a causal role, just not a conscious one - since they aren't conscious. The critical issue is that there's no basis for holding them accountable. (more on this later).

Without getting lost in the weeds on the causal aspect, could you elaborate on why you believe the AI's cannot be held accountable, and why do you think the human brain and nervous system is different than a cybernetic neural network? To me this sounds like a anthropic bias.

You don't need agency to HAVE compassion. You need agency to act on this compassion.

I disagree. Instincts are something everyone would agree is an automatic action, which require no agency to be instantiated. I am simply saying all action looks like instincts when you have enough information.

You're analyzing an instance of an optical illusion - which are notable only because they are exceptional. I'm talking about sensory input IN GENERAL. You don't skeptically analyze all the objects you encounter in the course of your everyday life simply because of the possibility you are misperceiving them.

It really is irrelevant if our reality is not composed of mostly optical illusions, although there can be an interesting conversation about how the brain is really constructing what you perceive, you are not really perceiving "out there," you are perceiving your brain's model. The reason your point is irrelevant, is because my argument is not based on the commonality of illusion, but rather, whether a particular thing actually is an illusion, which it seems obvious, that the phenomena of self, will, and consciousness, are all just mental constructs, not some spooky thing which floats to the left of your prefrontal cortex.

My position is that "free will" is a concept associated with responsibility and accountability.

Sure, that's what all compatibilists argue. I just think it is an unnecessary maneuver. A rapid dog has no "agency, or free will," but you would shoot it if it was attacking your baby. Invoking free will in order to have accountability is an artifact that is no longer needed.

It makes perfect sense to hold someone accountable for their actions: the action one takes are a consequence of one's beliefs, genetic dispositions, environmentally introduced dispositions, one's desires and aversions, the presence or absence of empathy, jealousy, anger, passion, love, and hatred. These factors are processed by the computer that is our mind to make a choice. If the consequences of that choice cause harm to someone else, how SHOULD others respond? Should they just excuse it because he had not choice (this seems to be the implication of your position)? No. We know he could have chosen differently had he been less reckless, or considered others, or any number of things. By doing so, that person becomes less likely to repeat the mistake - because he will have learned something. In effect, his programming will be changed because consequences provide a feedback loop that changes him.

You are just arguing against a straw man. I never made such an argument. I never once said I am advocating for undermining any notion of responding to someone who may be harmful to someone else.

Suppose the AIs in your example could experience pain, pleasure, regret, empathy, love, hate, and if it had desires that it worked to fulfill for the positive feelings it would experience, and aversions that it avoided because the negative feelings it would experience. Also suppose it could relate its choices to the consequences including the emotions it invoked, and that it could reprogram itself so that future choices would produce more positive and less negative outcomes.That would be closer akin to the "free willed" choices of humans. Whether or not we call it "free will" is irrelevant - my point is that accountability and responsibility comprise a feedback loop that we should acknowledge exists, and be glad of it. You weaken or break the loop when you deny accountability.

This is just an extension of your learning argument, which I already responded to. I do not think learning something or having code or neural networks altered does anything for the notion of free will, which is why I invoked the AI example in the first place.

While I agree with what you say, it is indeed possible that consciousness is indeed how you describe, I would simply say the reason consciousness appears as a problem, is because we are not recognizing that it is simply a trick our brain pulls, like connecting the dots of an imagine, or other wising rendering its model of reality. I would argue that deliberation is similar. Just the product of brain activity, a kind of post hoc connecting of the dots.

even under comptaibilism the choices are determined, so I do not see what is relevant or interesting about "free will." it is incoherent as a concept

— rlclauer

It makes perfect sense to hold someone accountable for their actions: the action one takes are a consequence of one's beliefs, genetic dispositions, environmentally introduced dispositions, one's desires and aversions, the presence or absence of empathy, jealousy, anger, passion, love, and hatred. These factors are processed by the computer that is our mind to make a choice. If the consequences of that choice cause harm to someone else, how SHOULD others respond? Should they just excuse it because he had not choice (this seems to be the implication of your position)? No. We know he could have chosen differently had he been less reckless, or considered others, or any number of things. By doing so, that person becomes less likely to repeat the mistake - because he will have learned something. In effect, his programming will be changed because consequences provide a feedback loop that changes him. — Relativist

You are just arguing against a straw man. I never made such an argument. I never once said I am
I wasn't arguing against you, I was trying to get you to think in terms of accountability. Set aside all concepts of "free will". Do you agree with my model of accountability?

I disagree. Instincts are something everyone would agree is an automatic action, which require no agency to be instantiated. I am simply saying all action looks like instincts when you have enough information. — rlclauer

Let's assume an action is, in fact, free. How would you tell from the outside?

The reason your point is irrelevant, is because my argument is not based on the commonality of illusion, but rather, whether a particular thing actually is an illusion, which it seems obvious, that the phenomena of self, will, and consciousness, are all just mental constructs, not some spooky thing which floats to the left of your prefrontal cortex. — rlclauer

But if everything is constructed, it makes no sense to call one construct "illusion".

Sure, that's what all compatibilists argue. I just think it is an unnecessary maneuver. A rapid dog has no "agency, or free will," but you would shoot it if it was attacking your baby. Invoking free will in order to have accountability is an artifact that is no longer needed. — rlclauer

Perhaps the point of accountability is to establish what you are not accountable for? After all, if we were to just eliminate possible causes of danger, we'd never stop.

You know...I always thought that, too. But then I came across this “degrees of freedom” for showing coordinate dimensions in a phase space, and I got to wondering how freedom was meant to apply there. I don’t consider freedom to be physical either, but apparently, somebody figured degrees of it, are. — Mww

Wouldn't you need a conception of physical freedom before you can map degrees of it?

One way to try to support free will is to attack the strongholds of determinism, such as the causal closure of the physical, particularly where we, as consciousnesses, seem to influence bodily behavior. But this doesn't yet speak to the freedom of the will. Actually, I am not so sure about that. Anyway, for free will to not be the case, it seems to me that consciousness must then be ineffectual. And if consciousness is ineffectual, why does it exist? More to the point, if it were ineffectual, how could we possibly even know about it?!!

Consider that last point carefully. If consciousness has no effect on behavior, how could behavior ever come to contain information about it or refer to it? When I tell you that I find myself to be conscious, it would seem that the fact of my being conscious has necessarily somehow made a difference in my behavior. It goes deeper still. For me to even form thoughts about my consciousness, for my consciousness to refer to itself, for me to know that I am conscious, my consciousness must have some influence on the brain state that presumably determines the structure of my mental state, which contains references to my consciousness. I am aware of my awareness, and my internal verbal and bodily gestural behavior often reflect this.

Epiphenomenalism, one of the classic determinist positions, would seem ruled out, as it is a belief in an inefficacious subjective something that we nevertheless find objective traces of in behavior. We would have a hard time explaining why the idea of epiphenomenalism ever arose in the first place if it were true. Bodies are referring to consciousness. It must not be the epiphenomenal consciousness that they are referring to!

You'll find people who assert epiphenomenalism, especially as they do implicitly when they take Libet's results as proof against free will, and then they will also say things about why and how consciousness evolved by natural selection because it offers some kind of survival advantage. Clearly, they haven't thought things through!


We might model epiphenomenalism as follows, with mental states designated "M" and physical brain states as "P".


Here, mental states don't influence physical states, nor do they influence other mental states. There is no way for any feature of the mental state distinct from the pure physical state to make its way into the next physical state. In other words, they are pointless and non-detectable. To wipe that top mental series away completely would make no detectable difference in the world. So why do epiphenomenalists believe in their consciousness? I think it is obvious that we should throw this stupidity out! Epiphenomenalists are a weird sort of dualist anyway, often without realizing it, even though they sometimes heap scorn on dualism. Honestly, I don't think a weaker position on the mind-body problem exists than epiphenomenalism. How it persists is beyond me.

For us to know about and to talk about our mental states seems to require that consciousness is efficacious somehow. Further, even if a physical state influences a mental state, a mental state must introduce something extra into the next physical state, rather than just mediating the physical-to-physical influence. This would mean that to at least some extent, the mental is free from prior physical determination. So perhaps here we have something of the will and its freedom. Of course though, it could be that there is some other sort of unbroken chain of mental causation that eliminates the freedom here. To have real freedom would seem to require that the mental is causa sui. "I determine while not being determined by anything outside myself. I am the fundamental substance."



When we talk about free will, aren't we basically talking about our minds in relation to our bodies? I think so. Doing something unconsciously (presumably automatically?) seems not to involve free will. So consciousness is an issue here, as is time. And these, it seems, are deeply connected.

I think when determinism is introduced, a lot of the things we blame or praise people for now will just be considered as outputs of deterministic inputs. The part of your model I disagree with is

hould they just excuse it because he had not choice (this seems to be the implication of your position)? No. We know he could have chosen differently had he been less reckless, or considered others, or any number of things.

Adding, "could have chosen differently," just muddies the water. should we promote good outcomes and mitigate bad? yes, but agency is irrelevant to that. In fact, I would go one step further. I would argue the idea of agency allows us to come to conclusions which are far less moral. For example, if someone is set up for failure by having a terrible childhood, their ACE score is 8/10, and they end up homeless standing on a street corner screaming at passing by cars, the notion of agency causes people to say, "well what's wrong with that guy? why doesn't he just choose to get a job or choose to stop drinking?" people are actually less sympathetic because we ascribe agency to people.

Let's assume an action is, in fact, free. How would you tell from the outside?

There are no "free actions." if you want to define free as uncoerced, which is a loaded word, fine. There is not a guy with a gun to my head. But this does not capture the idea of the causal chain of material factors which generate what we perceive as "conscious deliberation." Consider the Libet experiment and the "pantyhose experiment."

But if everything is constructed, it makes no sense to call one construct "illusion".

It makes sense to call something an illusion because of the disconnect between how it actually is and how it is perceived. The only "constructed" element here is the perception, not the real.

Perhaps the point of accountability is to establish what you are not accountable for? After all, if we were to just eliminate possible causes of danger, we'd never stop.

I do not disagree with that. I think it is usually framed positively, (what one is accountable for) because of our commitment to agent causation.

I erred by saying "could have chosen differently". Do you agree with what I wrote if I change that section to:

Should they just excuse it because he had not choice (this seems to be the implication of your position)? No. We know have strong reasons to believe he could would have chosen differently had he been less reckless, or considered others, or any number of things.

I agree with most of the rest of your comments (not your rejection of "agency", but I'll set that aside for now). What I infer is that there are cases in which the external causes (e.g. poverty, terrible childhood...) are mitigating factors. Those are important, but I don't think they imply we should abandon the notion of accountability I outlined. Here's a real-world example:

A few years ago, a 40 year-old woman in Houston was driving home late at night following a night of drinking at a bar. She got a bit lost, and wound up on an unfamiliar, dark street where she ran into a bicyclist who later died. The driver did not stop, but drove to her home. She was convicted of "failure to stop and render aid". In the punishment phase of her trial, her lawyer was pleading for leniency because she'd had a tough childhood: she came from a broken home, her single,twice-divorced mother was an alcoholic who never disciplined any of her kids, but she'd react at their bad behavior with verbal abuse. The mother kicked her out of the house when she was 14 because the girl was wild (promiscuous, using drugs, shoplifting). She and 2 of her 3 siblings dropped out of high school. Her brother is also in prison for repeated DUI (fortunately, he hasn't killed anyone). This was not an impoverished home - they were upper middle class.

There's a clear causal relationship between the girl's childhood and her recklessness at driving drunk, and failure to take responsibility for her actions. However, there are very good reasons to hold her accountable. Maybe, just maybe she will be more responsible when she's released from prison. Maybe the publicity this received will, in some small way, induce some people to take more care. Had she been given a slap on the wrist and sent home, it would have been only a matter of time before she was again driving drunk and risking lives. It is unfortunate that the girl had such a horrible childhood and suffered for it, but it's even more unfortunate that a completely innocent bicyclist was killed.

Wouldn't you need a conception of physical freedom before you can map degrees of it? — Echarmion

I would think. But in some defined coordinate system containing an observable object, the freedom it has is subject to the laws of the system, and the degrees of freedom it has are restricted by the system itself. Pretty lame kind of freedom, is all I’m sayin’.

Typical, though. Assigning a concept where it seemingly has no business. If we both hold that freedom isn’t physical, is that the same as saying freedom doesn’t belong to physical systems? And if that holds, why would there be such a thing as degrees of freedom in a physical system?

Well, I do not think there is a problem with a punitive course of action in the case of the drunk driver. Just like the AI which learned Go, learned by playing itself, no agency, no consciousness. We can introduce information into the system which alters the components of the system (individual brains), and the notion of agency, is completely unnecessary for this. In fact, as I argued earlier, agency obfuscates the reality, in a terrible way.

Take the case you provided. You can look at her as an agent, who because of her moral depravity and "evil" nature, decided, after conscious deliberation to get drunk and then drive. It would be easy for someone to conclude, "I hate that miserable person, look at the harm they caused!" Exactly as Sam Harris argued, the logic of hate dissolves when agency is scrutinized through the lens of cause and effect.

Your story illustrates that the capacity for compassion can be realized in the case of knowing the causal chain of behavioral production. If people knew the back story, they would probably be a bit more sympathetic, although overcoming the horror of someone getting run over (especially by a drunk driver - which most people think of consuming alcohol as 100% a choice, whereas, if someone had a stroke while driving and hit someone, they would not be horrified), is almost impossible in this case.

I can appreciate the case you laid out, because it attempts to show that some things are within control and some things not, while placing these in the context of moral responsibility. I just think yes, punishing a drunk driver is fine, because they can be dangerous. However, assuming there is a ghost in the machine driving the behavior is worse in my opinion, because it inflames that aspect of human psychology associated with retributive justice. Recognizing the lack of agency lessens that sting, so to speak.

To me it sounds like your motivation for preserving agency is its utility in punishment. In my opinion, this is not needed. I probably have not convinced you of that, but I do not think there is justification of invoking this agency, when it is a spurious notion to begin with, and it is not needed to establish order within a criminal justice sense. Therefore, we do not need to invent a God to prevent people from doing bad things, we do need to invent a free will, or a causal agent to have order within society, and separation of peaceful living and dangerous chaotic elements.

We have strong reasons to believe he would have chosen differently had he been less reckless, or considered others, or any number of things. — Relativist

It is that if his will was different than it actually was, he no longer being him, so to speak, the choices may have been different.

We have strong reasons to believe he would have chosen differently had he been less reckless, or considered others, or any number of things. — Relativist


It is that if his will was different than it actually was, he no longer being him, so to speak, the choices may have been different. — PoeticUniverse

Determinism does not entail incorrigibility. You become a different "you" when you learn something. A person can learn from his mistakes; he can learn that there are consequences. He can even learn to think more rationally, or coping skills for anger management.

I bring up agency to provide a locus for accountability/responsibility. and to highlight the difference between a causal chain and a chain of logic. The latter entails fatalism: the view there's nothing one can do to affect behavior (one's own, or anyone else's). We avoid fatalism by accepting that future behavior can be affected.

Determinism does not entail incorrigibility. You become a different "you" when you learn something. A person can learn from his mistakes; he can learn that there are consequences. He can even learn to think more rationally, or coping skills for anger management. — Relativist

:up:

:death:

:death:

There are no "free actions." if you want to define free as uncoerced, which is a loaded word, fine. There is not a guy with a gun to my head. But this does not capture the idea of the causal chain of material factors which generate what we perceive as "conscious deliberation." Consider the Libet experiment and the "pantyhose experiment." — rlclauer

Right. I am asking you to do a thought experiment though. Assume that, metaphysically, true freedom is real. How would it manifest physically?

It makes sense to call something an illusion because of the disconnect between how it actually is and how it is perceived. The only "constructed" element here is the perception, not the real. — rlclauer

But how do you know what's real? If both the internal perspective (freedom) and the external perspective (causality) are constructs, neither is real.

If we both hold that freedom isn’t physical, is that the same as saying freedom doesn’t belong to physical systems? — Mww

I'd think so, because to be part of a system, there'd have to be some kind of interaction. As system, the way I understand it, is a collection of factors with relevant interactions for a given question.

And if that holds, why would there be such a thing as degrees of freedom in a physical system? — Mww

There'd have to be degrees of interaction. So, maybe there could be degrees of determinism, and what remains without interactions is than negatively defined as "free".

:death:

:death:

Automatic" actions taken in habit and familiarity could be considered to be choices, but only in the sense that a person has accepted living in them. — Pathogen

How very reminiscent of Hume-ian British Enlightenment empiricism are you!!

Peoples choices (...) influenced by hormone signaling that is never perceived at a conscious level. — Pathogen

How very reminiscent of Skinner-esque empirical psychology are you!!!

It’s ok, though, no big deal; those are both....ehhhh, somewhat.......historically at least....valid subsets of their respective doctrines with respect to the human condition in general. My only point would have been that neither of those, while being sufficiently explanatory regarding “unconscious choice”, can have any bearing on arguments for or against free will, which requires necessarily its ground in a priori rationality, absent by definition in both of them.

Moving on.....

Allow me to try to express how I think our views differ within a narrative,

Two boats are traveling down a river side by side. On the first boat the captain thinks to himself, my first mate is an experienced sailor and knows how I run my ship, I'll let him direct it while I enjoy the breeze. The other boat's captain knows his first mate is well experienced, he could talk with him about his future plans and rest assured they would be carried out, but as a matter of preference the captain takes the wheel and as an action in that moment steers the ship while enjoying the breeze on his face.

Of course, both captains enjoy the breeze :wink: — Pathogen

In my view, just to clarify, I think most of us can be either captain depending on the situation. I personally tend towards the latter, while I recognise there are many who prefer the former. That’s part of the freedom...

It’s probably impossible to ever know why or how the phrase “degrees of freedom” was chosen to represent the observable conditions of a physical system. I surmise its author had something in mind, such that the freedom relates some object to its lawful probabilities, and degree relates that object to a range within those probabilities. Or....he just picked the notion out of a lab-coat pocket because it sounded awe-inspiring. Who cares, really. The point is, freedom is strictly a metaphysical domain, an immaterial/rational....transcendental..... notion meant to qualify its object as to that object’s possession of it, but never to quantify its object as to how much of it that object has.

Which leaves us with, in a human physical system, in what manner of speaking can your “freedom absolutely applies” be true?
1.) If the will is free to do this but not free to do that, then it follows necessarily that freedom, in and of itself, cannot be a condition of the will, but must be either logically removed or temporally displaced from it.
2.) Or, there is no such thing, which is immediately self-contradictory from a metaphysicalist’s point of view with respect to morality and from a physicalist’s point of view with respect to degrees of freedom.
3.) Or, and which is much more prevalent, freedom is applied as an empirical predicate in the human system, just as it is in the dimensional domain in observable coordinate systems. Case in point...I am free to move my arm, free to like vanilla, and other similarly constructed absurdities, having nothing whatsoever to do with the faculty of will.

Opinions and noses........everybody’s got ‘em.

I personally do not see a distinction between a chain of causation and a chain of logic. Fatalism may not be correct because of quantum indeterminacy. However, QI does nothing to support the notion of free will

Right. I am asking you to do a thought experiment though. Assume that, metaphysically, true freedom is real. How would it manifest physically?

It would manifest as an action free of prior causes. This is why I think free will (or true freedom as you put it) is an incoherent concept.

But how do you know what's real? If both the internal perspective (freedom) and the external perspective (causality) are constructs, neither is real.

You do not "know" what is real, if you can only experience it subjectively. If you do scientific testing, and continuously get the same result, you can conclude that there is probably a real property which is affecting this outcome. I do not really know what you mean by "internal perspective" vs "external perspective." In my opinion, there is your perspective, which is subjective and therefore fallible, and there is the world we inhabit, which seems to be real, and we have discerned some properties about this world, but the discovery of those properties requires placing a check on our subjectivity, namely the scientific method.

:death:

:death:

Here's what I was referring to:

Contrast chains of logic with causal chains. Where Pi are propositions, and Ci are causes:

P1 =>P2=>P3=>....=>Pn (P1 implies P2, P2 implies P3...)
C1 ->C2->C3->...>Cn (C1 directly causes C2, C2 directly causes C3....)

Logic is transitive, so the above logic chain entails: P1=>Pn for all n. The intermediate Pi are irrelevant.

On the other hand, causation is NOT transitive. C1 does not directly cause Cn for all n. No Ci in the causal chain is irrelevant.