How physicists handle the idea of Free Will?

pftest

But the thing is that "logical supervenience" is a mental construct. It has no physical meaning. This is not a useful concept for the actual physical relationship between consciousness and the brain.

Q_Goest

Hi pftest,
Logical supervenience is a concept like the terms "intrinsic" versus "extrinsic", and like those terms, it attempts to pick out a relationship between properties. In this case, logical supervenience tries to pick out properties that relate. So the argument is, "Do mental states logically supervene on physical states?"

For the record, logical supervenience is nicely defined here:
Let's then ask the question as Chalmers did, do mental states logically supervene on physical states? Could we for example, imagine a Turing machine that could consistently pass a Turing test that does not support mental states? That is, is it possible that such a machine could only have physical states? Note that mental and physical states here are defined as Jaegwon Kim defines them.

Clearly, a Turing machine produces responces based on algorithmic manipulations of symbols. Those algorithms can be described mathematically and are deterministic. So for each physical state through which a Turing machine passes, there is a physical reason for why it passes through that state. There is no need to appeal to mental states in this case in order to explain why that Turing machine produced the set of responces it did in order to pass the Turing test. Such a machine therefore does not need to have mental states in order to produce the behavior, it only requires the physical states. If we accept this, we can say those mental states to not logically supervene on the physical states since we clearly have no reason to suggest that subjective experiences (which can't be objectively measured) should supervene on those physical states, though we still might claim that those mental states naturally supervene on the physical states.

In fact, we can't know if there are ANY mental states that supervene on physical states if the mental states have no influence over the physical states. That problem is known as the Knowledge Paradox as described for example by Rosenburg and Shoemaker.

Goodison_Lad

Legal systems are based on the assumption that free will does exist and that people, generally, are not operating merely deterministically. People are considered to have genuine choice, not just the illusion of it, and the main reason prosecutions occur in the first place is that the accused are believed to have exercised said free will in a way the law disapproves of. It is the very notion of free will that underpins the concept of responsibility, and so people are accordingly held responsible for the consequences of their free choices (an expression which is tautological - what meaningful sense could be made of the idea of a ‘non-free choice’?)

The basis for mitigation is where it judged that free will has been significantly compromised, perhaps to the extent that it was completely absent – mental illness, intellectual impairment, duress etc. Society considers that responsibility and free will are two inseparable sides of the same coin.

So I really don’t think we can have it both ways on this point: we have to be consistent. If our actions are deterministic, we can no more be responsible for them than can a dog for barking, a brick for falling or a supernova for exploding. The fact that other automata might act as though they believe I can be held responsible is simply a further manifestation of their delusion. If they were to fully adopt the view that my behaviour is deterministic, and therefore a direct product of effects immune to non-deterministic intervention, they would have to conclude that the concept of responsibility is a redundant one. We might like it, but it is not logically justified.

However, even if the automata of society were to acknowledge that I had no real free choice, this would not mean that they should necessarily act differently. The imposition of a legal code on an automaton might have the effect of being another deterministic factor affecting its future behaviour.

If responsibility does exist without free will we’d have to consider the possibility of extending the legal system so that we could prosecute animals. Am I any the less a victim of determinism than a pigeon?

sophiecentaur

This is true, of course. Society evolves to encourage appropriate behaviour. Social pressures and laws have a strong influence on our behaviour (unless we are particularly deviant). There is a social 'system' that tells us we are 'accountable' (only a word) and that influences the way we are likely to behave. Evolution has made sure of that - but no more, in essence, than it governs the behaviour of other animals on a much simpler level.

But the fact that we feel we have a free choice in our actions need to be no more than, as I have said before, a way of rationalising what we have just done or 'decided' on. The illusion of free choice is, in fact, very necessary or we'd just lie back and let it all happen - in the belief that it isn't worth trying. The fact is that all life forms 'try' (i.e. behave as if they were trying). It is just the fact that we are so complex that we had to evolve a consciousness in order to handle it all - so we are 'aware' of trying.

madness

The external world is a mental construct, the self is a mental construct, I can't think of anything at all that isn't a mental construct.

Pythagorean

Er... no. That's exactly wrong. It only works because it's a deterministic system: because we can have a causative effect on people's actions. If there was no causation, people would make decisions independent of whether it caused them suffering. But because the causation chain is not broken, having and enforcing law continues to work.

Just like any other organism, we change our behavior in light of new information as long as that information passes a threshold in our emotional significance detectors (probably mostly in the amygdala and basal ganglia).

No, society doesn't care about free will, only responsibility. All society is, is a bunch of individual voices that want safety and security for themselves, so they don't like people engaging in risky behavior around them. It's really quite normal (especially in mammalian organisms) to have some kind of system that eliminates cheaters, for instance.

What we call mental illness, intellectual impairments, duress, etc, are all examples of when executive function is not dominating prediction and decision measures in the brain. When executive function is broken, people do not care about participating in social acceptance. For instance, frontal lobes complete wiring until between age 3-5, when toddler's start caring. The next major finalization comes in the early 20's with myelination, which finalizes the circuit dynamics in the frontal lobes. This gives the organism a long sample-time while the circuits are still plastic to figure out, negotiate, and even create new social rules and paradigms.

When we get frontal lobe damage or deterioration of any sort (whether from traumatic injury, disease, or other morphological or development abnormalities) we care less about what society thinks. The most famous case of this is Phineas Gage, but there have been countless examples since. Rather than seeing them as lacking free will, we can see these people as having broken their detector/predictor mechanisms for social instances.

I think it's best to distinguish between free will, which is a supernatural idea that some force acts independent of cause and effect, and will power, which is the ability for an organism to execute it's needs/wants (determined from biology and environment). Willpower definitely exists and it's how we judge responsibility.

If somebody always wants to kill you, they are responsible. But if somebody has a disease where there arm swings around wildly at random times, we don't consider them responsible. This is independent of whether the system is deterministic or not; it's only a matter of which system is dominating interactions in the organism.

madness

http://www.wjh.harvard.edu/~jgreene/...ilTrans-04.pdf

This looks very relevant to the discussion. From the abstract:

"New neuroscience will change the law, not by undermining its current assumptions,
but by transforming people’s moral intuitions about free will and responsibility. This change in moral outlook will result not from the discovery of crucial new facts or clever new arguments, but from a new appreciation of old arguments, bolstered by vivid new illustrations provided by cognitive neuroscience. We foresee, and recommend, a shift away from punishment aimed at retribution in favour of a more progressive, consequentialist approach to the criminal law."

sophiecentaur

I can't be sure what you mean by that. I don't actually believe in the supernatural (everything is 'natural' in my view) so are you saying that it 'really is' supernatural or that it's the interpretation that people give for certain experiences they have? If you mean the latter then I agree with your point.

Pythagorean

I'm saying that I think free will is a lot like a soul or a god when it comes to explaining mechanisms (i.e. its not much of an explanation). In science, we expect phenomena to follow rationally from known laws of physics (and it consistently does). Behavior of organisms (even humans) shouldn't violate our rational view, and all our experiments have shown it doesn't; we have a working model of brain function and neuroethology.

There is no need for a supernatural explanation. There's nothing in decision-making that requires explaining free will (though as someone previously said, there is probably plenty of valid interest in studying why we experience the sensation of free will).

Goodison_Lad

My point wasn’t whether legal systems are effective because they assume free will. They do assume it, and it’s not why they might work. I did actually acknowledge in my last post that their efficacy is based on their potential to have a causal effect on the supposedly deterministic system that is a person.

I must admit to being a bit baffled here. I know of no relevant system within a society that does not make the link between the responsibility of an individual and that individual’s capacity to make free choice – it is almost axiomatic. This assumption may very well be the representation of evolved strategies to minimise danger, but the net result is that people attach blame to wrongdoing because of the belief that alternative courses of action were open to the criminal. There are people who do not buy into this – they conclude that all behaviour is beyond the control of the organism, ‘control’ implying the ability to have done other than what is eventually done. For them, no control exists. Such people would be being true to the deterministic principle.

The burglar receives our disapproval precisely because we assume he didn’t have to enter our houses – he chose to. If we thought he had no real choice available to him, why should we hold him responsible for his behaviour? Only when we allow for mitigating circumstances do we absolve, to a greater or lesser extent, the criminal from his behaviour. Of course, some people won’t allow for any mitigation: the criminal made his bed, he can lie in it. Choice.

I don’t see at all how will power justifies the introduction of responsibility. As you describe it, it is a deterministic phenomenon, so the organism can have no real choice but to execute the orders of its will power. The brain of the organism will analyse the information and presumably generate what that brain considers to be the response needed to fulfil the organism’s needs/wants. At no point need this process be elevated above the automatic to explain it. Even if an apparent range of options is available to the executive function of the organism, the one it ‘chooses’ can only be the result of an automatically running programme. There is no choice in will power.

The only way I can see your point is if we mean different things by the word ‘responsibility’. If it is taken to mean that the fox is responsible for the slaughter in the chicken shed because we can attribute the slaughter to the fox’s behaviour, and no more than that, then yes, the fox is responsible. But in the way I mean it (and, I suspect, the way most people use it), responsibility has a moral dimension over and above a simple attribution of effect to cause.

Determinism dictates that the person who wants to kill me, in executing the will power programme, is powerless to do otherwise. Should there be an element of genuine choice as to whether to execute the will power programme, this would have to be non-deterministic i.e. free will.

Determinists must accept, er, responsibility for their beliefs. They can’t have their cake and eat it.

Goodison_Lad

Thanks, good article. It both supports and goes against my argument in parts!

The assignment of responsibility is taken as a pragmatic step, not a moral one - by allowing the law to do this and act accordingly, we get a better society.

Morally, you can't hold the criminal responsible because determinism says he has no free will; pragmatically, you can say you want to hold him responsible because this helps you do things that can affect his future behaviour.

But you don't really believe he is truly, ultimately responsible. Just don't tell him that!

Greg Bernhardt

Not sure how exactly relevant this link is but I heard this on NPR today

The Subliminal Self
http://audio.wbur.org/storage/2012/0...minal-self.mp3

sophiecentaur

I'm not sure that follows, strictly. We are observing this 'game' from inside it and could well be interpreting such scenarios in a too - simplistic way. If I want to kill someone but then restrain myself then the outcome could be just as 'determined' as if I did actually kill them. I might feel that I had made a decision but how can I be sure of all the factors that the decision algorithm had used? How can I be sure that I 'made a decision' independently?
'Responsibility' could just be a way for society to impose its influence on us. Our evolution has included what is advantageous to the species as a whole (and, indeed, to other species).

I have concluded that 'god' has been invented (as part of our evolution) in order to get individuals to behave 'better' to other humans. The same could be said about conscience and consciousness. Where that all leaves us is a bit problematical but I must say, it hasn't stopped my enjoyment and appreciation of life. And it certainly is fun to discuss.

Pythagorean

Ok, well first off... let's just get one thing straight. The law are not neuroscientists, so they're not exactly an authority. Neither is society at large. It's really a complicated subject; it will take several iterations of posts clarifying the language and the ideas.

But yes, that's the difference. You've built a notion of free will into your definition of responsibility. So let's skip that semantic argument.

We still hold the burglar accountable. I don't personally care whether he did it deterministically or not. I don't want to punish people for retribution. I just want dangerous people off the streets so they can't be dangerous. It's not a personal thing, I don't want to see them suffer. They're just dangerous (or costly) so their risky behavior shouldn't take place around me or my family. They can perform risky behavior in a place designed for it (a prison). But we know, statistically, that the burglar probably didn't get a good education or have a stable family life growing up. So we launch programs for prevention. This all relies on the system being deterministic.

To you first paragraph posted, this is exactly what the evidence shows. You give people more information, they make better choices (look at the correlation between crime and education and crime and social class). People will always make the choice they perceive as the best choice as long as their executive functions are working properly. If you have more information (education) and money (resource), then you can more easily carry out the best options.

now IF your executive functions aren't working properly, then we don't hold people accountable. We acknowledge that some part of them is broken, so throwing education and money at the problem won't fix it; that's an important distinction so that we don't waste resources (but then again, humanitarians probably won't let you be picky like this).

Of course, everybody has a different genetic concoction underlying this all, so many different biological responses can come from the same environmental stimulus. The developmental period in the womb is important too. Monozygotic twins aren't EXACTLY the same due to some mutations, but also, importantly, due to different nutritional paradigms in the womb (depending on who's closest to mama's nutrition source).

Also, what Sophie said.

Pythagorean

Actually, what sophie said is grounded in neuroscience (an internal competition between brain regions). That's one of the well-known jobs of the frontal lobes. They have inhibitory projections to the rest of the brain, acting as a break. The "rest of the brain" has a lot of reptilian parts in it that initiate seeking behavior (basil ganglia) or react emotionally to stimuli (amygdala)

The frontal lobes are compromised when you drink alcohol. As I said before, toddlers make a transition from being disobedient little tyrants to listening better around age 3-5, when their frontal lobes go through a wiring paradigm. Then lastly, the frontal lobes myelinate in the early 20's, when kids stop being teenagers (right about when we let them drink alcohol).

sophiecentaur

This is what annoys me so much about 'those danged Philosophers'. They are just not in a position to go back to square one (because they are inside the system'. They go back to a point that, somehow, satisfies them as being far enough back - then they build whole edifices on that point and reckon they've actually proved something.

Problem is that some of them are extremely clever and they do deserve some recognition but, in the end, they are basing all of their stuff on the dreaded 'faith' word (even the atheists).

I guess I'm just an old fashioned Utilitarian.

sophiecentaur

Perhaps the problem that people have with just accepting their machine-like nature is that we also have evolved with a need to feel that we're something special. It wouldn't be favourable if everyone thought of themselves as being without a 'soul'. For a start, people would feel a lot better about killing other people if the general opinion was that they didn't have souls. But I don't believe we do - so what do I do? I have to use my Intellect instead and go with my emotions (an electro/chemical mix). Never felt like killing anyone despite that. Is that really worse than having a religion?