Yes, it's a brilliant book that proves what we see as free will is obviously not free and it's not even up for contention that we actually have the ability to decide, even though it feels as though we do.
This appears to be a bit of bubblegum neuroscience. First, the article states:
He asked participants to report, using the clock,
exactly the point when they made the decision to
move.
Then, it goes on to say:
There’s no reason to think that we are reliable
reporters of every aspect of our minds.
Well, this is quite the conundrum. Which is it? Report exactly when you do something or accept that we are not "reliable reporters?" Yet the author goes on to further say:
Even supporters of Libet have to admit that
the situation used in the experiment may be too
artificial to be a direct model of real everyday
choices.
Given the delays in perception we all experience due to working at "chemical speed", as opposed to light speed, this experiment smacks of fast-twitch muscle measurement more so than exercising free will.
Even the original premise of Libet's experiment negates a free will choice:
“let the urge [to move] appear on its own at any time without any pre-planning or concentration on when to act”
How is this free will? Free will is a conscious choice to do something, not a subconscious act.
It can also be a case there's a cognitive delay associated with taking note of the time. The mechanism involved into decision making for muscular movements might be dramatically faster than cortical perception and verbalization of time.
The experiment seems to boil down to, "Processing and propagating signals encoding decisions, to report them to the executive centers where the conscious mind will notice them, takes time."
That's not really a wild statement about free will. It's just another paper showing that brains are made of atoms.
Really I've seen so many terrible psychological experiments that spawned huge articles that I am not taking any of this seriously anymore. I don't doubt that we will get somewhere at some point, but it's very clear that at this point we are mostly bluffing.
Just because you set some kind of experiment that turns up some kind of data shouldn't mean that you compare your studies with science on the level of physics or astronomy. Unless your experiments provide a model that has any kind of predictive power you are probably getting nowhere.
The scientific method got where it got, because it started with people doing small experiments that turned up to generate incredible models with a great predictive power, and started building up from there. Just copying the method, disregarding the predictive aspects, and just bluffing from there to generate nice sounding headlines is hardly a similar path.
Libet's experiments were carefully planned and are highly regarded. That's why they have never been refuted, and in fact spawned a lot of newer studies in the field.
They can be valid experiments that reveal important insights about the workings of human consciousness and still be utterly irrelevant to the question of free will. With 100% certainty, because this is the case.
Much of the confusion about free will and how much free is our will comes from experiments that study limit cases like our body reactions to external stimuli or something of the sort.
But I don't think anyone can seriously state that we are absolutely free, specially when our bodies are involved. Extreme cases reveal that that there's much to say about how we (body+mind) work, and we're certainly not absolutely free.
The simplest and, perhaps, best argument for our partial freedom is that we can ask questions. I don't think there is any kind of property in the physical world that would yield questions and truth-intuition about answers to those questions. Electrons just bump into each other. They don't ask questions about it. They move on like there's no tomorrow.
Software which clearly does not have free will asked a question allowing you to enter that post.
So, if you exclude the language part of asking questions as cruft the type of experiments that human baby's do is the same type of experemnts birds do when learning to fly. But, biological examples of this get really simple and plenty of people have built non biological examples.
At a lower level you can argue that QM shows every time a wave function collapsed a question was asked and answered.
Which is really the issue of free will, most definions boil down to either, 'humans' or include simple systems that clearly lack choice.
You could think of that software's question as an extension of the creator/owner of the software. Said person is asking for your opinion and delegating to the software the ability to receive your opinion and display it to other people.
By that notion, by opening or closing a door or even standing on some sand I endow it with free will. Which is an odd way of looking at the world as more or less everything on the earth would then be endowed with free will.
Honestly, I can see that what you are taking about being a property. But calling it free will just muddles the debate.
Random implies no agent (self) at all. It's just a roll of the dice.
Pre-determined, on the other hand, it's not incompatible with free will (precisely because free-will doesn't mean "randomness" -- it just means it's not imposed upon you by an external agent.
Let's imagine a totally deterministic universe. How could ever pre-determined play well with "not imposed upon you by an external agent"?
Because this pre-determined you is not an external agent imposing its will on a third subject, it's exactly who you are as a "self".
That is, you are your history (and JUST that), and the only decision your self could make is a decision based on its precise history, which is exactly what it makes every time.
Anything else wouldn't be YOU or YOUR free will, it will be arbitrary (random).
So, I think people just have this misconception that "free will" should have the emphasis on "free" -- to mean "whatever".
But the emphasis should be that it's a "will", i.e. the expression of the subject. And such an expression can only be understood in the sense of a subject that has history (you weren't just born this instant) and that makes decisions precisely drawing upon that history -- not just as a source of inspiration of his final decision, but as the total embodiment of his past (with all its aspects) in his decision.
In other words, free will is not you being free to will anything. It's just you being free to be you -- an entity of a concrete past.
I disagree with at least one of your premises. In a deterministic universe, one could (assuming perfect knowledge) predict this exact comment you would make now, thousands of years before you were even born. How is that free will?
I disagree that there can be free will when all your actions can be known to an omniscient party beforehand. Where's the freedom in that?
>I disagree with at least one of your premises. In a deterministic universe, one could (assuming perfect knowledge) predict this exact comment you would make now, thousands of years before you were even born. How is that free will?
How is that not?
Your objection just restates "for me free will shouldn't be deterministic and its decisions shouldn't be known in advance".
Whereas the core of my argument was that it doesn't matter if it's deterministic and known in advance, because it's still the inevitable expression of one's self (which has a certain history which makes it itself).
What you're saying would imply the need not just for a self, but for said self to be independent of your history, like an external agent, and able to make decisions that are not depended on its past (and hence its presice state).
But (my argument is) the core and sole essense of one's self is it's past (as expressed in your state this very moment). That, and nothing else.
Note also that traditionally "God knowing everything" in advance didn't sound conflicting with free will. Free does't necessarily matter that the decision has to be taken on the spot -- just that it will reflect what you want it to be.
Put another way: you StavrosK has to take a decision now.
What would "free will" be in this case?
Some thinking thing (your soul?) standing above your brain and making decisions that might or might not fit with your past, brain wiring, memories etc?
Or you (your brain wiring etc) making a decision by drawing upon its very state, which in turn is affected by its past experiences and previous history, to the very detail, so that it can only make a single specific desicision (the one you end up taking)?
Where would any other decision besides the one you take come from? That would be like a third person being inside your head, given all the details and told to make a decision that might be whatever...
P.S. Not also that determinism != ability to predict in advance necessarily, even if you have perfect knowledge. For some systems (e.g. certain finite automata) you have to run them and you get the state piecemeal, by running them only.
There's no shortcut function that tells you the state at moment X(10000) without having to do through the previous 10000 steps (unlike say ballistics, where if you know the initial power, angle etc, you can calculate in O(1) the path at any later moment).
Of course you could always run the steps faster. But if the universe is deterministic, then you'd have to run the universe faster, which is not doable.
(This argument or similar is put forward by Wolfram IIRC).
We're getting deep into philosophy here, but I'm not sure that, if you froze the state of the universe right before someone made a decision and replayed it, they'd make the same decision every time. I guess that's the whole difference.
What would have changed in the case of the replay though?
For something to change it either needs randomness (which itself is not "free will") or an agent independent of his material status.
Not just independent from his body etc mind you: but also independent from his neurons, neural connections, memories stored in the brain's substrate etc.
Else, if all those are the same, shouldn't the decision be two?
I've heard two alternative solutions to this questions:
One from Penrose, who implies that quantum mechanic mechanisms in the brain (microtubules) bring randomness and casual independence into it (but it doesn't strike me as satisfactory, because randomness it not free will, and besides the theory has had some hard criticism).
The second is the classic idea of soul -- which is independent from the body and can make decisions outside of the constraints of its physical status (e.g. if it was "replayed" as you say, it could have made another decision).
This is not very satisfactory to me either, not because it assumes a metaphysical soul (I'm willing to accept that), but because it assumes one's self (his "soul") is independent from his past, and given the same exact past experiences it can come to two different conclusions (when "replayed" etc).
But if we aren't our past, and the way our memories and brain matter is wired, what are why? Who would that soul connect to my inner being, if it's independent of my past?
The same can be said for the specific arrangement of dust interstellar gas. Suggesting said gas has free will just means your definition of free will is basically matter with a past or even more abstractly just math.
>The same can be said for the specific arrangement of dust interstellar gas. Suggesting said gas has free will just means your definition of free will is basically matter with a past or even more abstractly just math.
Yes, the same can be said for it. Though, obviously will from will can have quantitative differences.
One involves much more complex math.
What else did you expect? God and an immortal soul?
What else did you expect? God and an immortal soul?
No, but if your going to say rocks have free will your using the term to mean something well outside the generally accepted understanding. At which point you probably want a new word to avoid confusion.
What exactly is the middle ground between randomness and determinism?
Pre-determined, on the other hand, it's not incompatible with free will (precisely because free-will doesn't mean "randomness" -- it just means it's not imposed upon you by an external agent.
[...]
That is, you are your history (and JUST that), and the only decision your self could make is a decision based on its precise history, which is exactly what it makes every time.
To me this seems to just push the problem a bit around. How does your self evolve over time? How does it make decisions? Is it deterministic or random? Or again what could a middle ground be?
About the experiment, how can Libet conclusively derive the decision came after the initial EEG measurement. I mean how did he know the duration it takes to report the decision. The origin of the decision to being able to reflect on it may take a variable amount of time therefore invalidating the claim.
Also the strict separation of conscious/unconsciousness seems simplistic. Most actions may very well be processed subconsciously with very little decision making bubbling up to consciousness. Driving a car comes to mind. This also means when asking a subject to process something impulsively the vast majority of their actions may be subconscious.
tl;dr
People spacing out does not invalidate free will.
The subjects reported the position of the dot, that's how he knew the exact time point in which they were 'aware' of their impending action. There was a comfortable time difference between the two.
These studies do not invalidate free will, they just show that free will is not something ethereal, but a product of the specific makeup of ourselves, which is encoded in our brains.
There is no need for these experiments: There is no free will, and you can understand this by simply applying pure logic/rational thinking:
When you're born (the "1st second" of your life) you're given 2 variables: (1) Your genome and (2) the environment you're born into. You have absolutely no control whatsoever over these 2 variables (with their infinite number of sub-variables they "contain"). And everything else that follows (the "2nd, 3rd, nth... second" of your life) is a function of these 2 initial variables: And this includes your brain that you will use to make the infinite number of choices in your life (both consciously, such as "I'm going to buy milk" and subconsciously, such as "firing neuron X at second Y", "moving atom A to location B", etc.).
EDIT: Downvoters - please state why this would be wrong.
You are right that it was an unnecessary comment, but declaring that "There is no need for these experiments" is a little armchairy. This is a debate that has been going on for a long time and a version which takes place on hacker news would be more useful if it wasn't immediately polarized.
Where you're correct is: We can never prove the absence of free will, because we can never understand then entirety of our environment (how the universe "really works", etc.). But when you take what we know today and apply rational thinking, it certainly looks like there is no free will.
> We all have the ability to act at our own discretion.
That is exactly what my argument would deny. There seems to be not even an inch of free will.
> Does it really matter
100% percent yes, for example: The absence of free will forces us to stop judging people, simply because the concept of "guilt" (on the negative side) and the concept of "merit" (on the positive side) are proven to be entirely baseless. If we applied this to our culture, we'd live in a totally different world.
I didn't mean to say anything about collective guilt or merit, but apparently I didn't communicate well; sorry.
In iterated PD competitions, it's clear that none of the programs involved have free will, but punishing defections is still an important part of either a strategy for getting ahead, or a strategy for reducing defections. That is, the programs still have to keep track of the guilt or merit of their peers in order to make decisions about how to react to past cooperation or defection. Regardless of the moral responsibility of humans for their actions, they still have a responsibility in historical fact.
>100% percent yes, for example: The absence of free will forces us to stop judging people, simply because the concept of "guilt" (on the negative side) and the concept of "merit" (on the positive side) are proven to be entirely baseless. If we applied this to our culture, we'd live in a totally different world.
Not really, it doesn't matter. If there's no "free will", then we have no say on whether we judge people or not, not even as to whether we believe in free will or no.
Your answer works accidentally on two conflicting levels, assuming that if we discovered that we have no free will we'd still have free will to act on a certain way upon it (not to mention that the discovery itself wouldn't be on our own free will).
>No because we'd actually act upon it because of purely logical chains (1) of cause and effect, not free will
First of all, the discovery itself wouldn't be an action of "free will".
Second, who said we can act on "purely logical chains of cause and effect" when there's no free will? Whether we act on those chains or not will already be predetermined by the "no free will" mechanism determining our actions.
If anything, as societies we act pretty illogical a lot of the time (heck, even most), for stuff scientifically known to be BS.
> The absence of free will forces us to stop judging people..
Why would that be?
If A harms B, then your argument is that B has no right to retaliate it, because, that A should harm B was inevitable, and was totally out of the control of A.
But my argument is that just like A's harming B was inevitable, so is B's judgement of A and the further retaliation in response to the A's action..Inevitable.
In other words, the world should just go on as it is, even in the total absence of free will. What is important, I think, like many other things, is the 'appearance' of having free will. Because without that, all life looses it's meaning.
> If A harms B, then your argument is that B has no right to retaliate it
Yes and no. Yes, because no judgment allowed means no punishment allowed. But: the correct answer to A's action would be a sanction against A, in the sense that it would prevent A from harming again in the future. The wrong answers are: judgment, retaliation, punishment, revenge.
> In other words, the world should just go on as it is
It will. And it's also exactly what's happening here.
> What is important, I think, like many other things, is the 'appearance' of having free will. Because without that, all life looses it's meaning.
Not to worry, we'll always have this appearance, because causes and effects work on every macro and micro level. No system of sensors will ever be able to measure all variable, no model will ever be able to include all variables and no computer will ever be powerful enough to model everything. So, the magic remains.
I think you both essentially agree. The sentence you quoted makes no sense in a world without free will but that is probably just an unintentional small mistake by the author because we are so used to assuming we have free will in our everyday life even if we actually believe we don't. The sentence just tries to say what we should do (assuming we had the free will to do so) if we discovered that there is no free will in our universe (which of course is contradictory). You could (probably) get rid of the contradiction if you understand the sentence as someone in a universe with free will talking about making changes to a universe without free will from the outside, like a god fixing an issue in one of his universes to make it more consistent.
If everything we do is predetermined from the instant of our conception — actually much further back than that since all events leading up to one's conception were also predetermined — then we are all going to still do what we do, which includes judging others by their actions. If some choose not to apply guilt or blame to others, well, then that was also in the cards.
If there is some variable or factor outside our measurable environment that provides for truly free will, then it does make sense for people to be judged based on their actions and for guilt and blame to be applied in appropriate cases.
Applied more generally, if there is no free will then whether an individual believes there is free will does not matter; the deterministic environment decides that matter. If there is free will, then there is a distinct advantage to understanding and believing that fact. So, assuming there is free will has a neutral and upside state, while buying into determinism has a neutral and downside state. Isn't the rational and safe position to just assume there is free will?
> If some choose not to apply guilt or blame to others, well, then that was also in the cards.
Correct. Like this discussion and everything logically was also "in the cards". So, yes, in a sense "it does not matter", while it still does.
> If there is free will, then there is a distinct advantage to understanding and believing that fact.
The issue is that you see an "advantage" in free will. I suppose you mean that free will would make you more powerful because you would not believe that you "can't change things", etc. This belief however is false. It's not logical. Even if we understand that there is no free will, we are still going to make "our own decisions" based on all the variables in our lives. The vastness of the variables in our networks is such that it can never be measured and modeled entirely. It is simply too complex for us to understand, which means that it will never feel like a network of causes and effects. So our choices will always feel like our own, and that is good enough for us, it seems. For example, I'm still going to try to create useful products and believe I can become successful, because that's what motivates me, that's part of my "variables".
The determinism kind of removes the "free" from free will. Determinism means that, if someone else knows your internal state, they can predict, or worse, manipulate your free will. People are naturally extremely personal and possessive about their free will, and these experiments are "stealing" it away. That's why the knee jerk reaction to these experiments is always extremely negative.
1) Free will is the capacity to act at our own discretion, independent of our current surrounding and context, including the physical state of our body and brain
2) Free will is the capacity to act at our own discretion, independent of our current surroundings and context, not including the physical state of our body and brain
Note: In both cases I'm assuming there is no soul, mind beyond the body, etc.
If you define free will as (1), then there is no free will, since any decision will always be subject to the physical state of my body and brain (except, of course, if there's something not physical in play as described in my note).
If you define free will as (2), then there is free will, since despite the context surrounding you, you can still make decisions independent of it.
The same could be applied to a state-machine (finite or not):
1) Free will is the capacity of a machine to produce outputs independent of it's current internal states and the external inputs
2) Free will is the capacity of a machine to produce outputs independent of it's external inputs
(1) means no state-machines have free will, (2) means they can have free will.
>If you define free will as (2), then there is free will, since despite the context surrounding you, you can still make decisions independent of it.
How do you come to that? Obviously the state of our current surroundings and past histories determines the state of our brain and body (down to the firing of the very last neuron).
> Obviously the state of our current surroundings and past histories determines the state of our brain and body (down to the firing of the very last neuron).
That's why I only said "current surrounding and context", I didn't include past histories.
Look at the machine-state example, it makes things more clear.
The issue with (2) is that the physical state of our body and brain is a function of all past and current surroundings and context (including our genome). Which means it can never be independent, but is instead fully defined.
This assumes that you were capable of conscious decision making, or had free will, at birth. The assumption is the problem, I think. I would think that "free will" is developed, not de facto available from birth. In fact, I think free will in a baby could be a pretty bad evolutionary disadvantage, whereas it would be a great advantage for an adult.
If you see life as a 4-dimensional network of variables linked through "cause and effect", then there is no place for a "free" variable which could introduce free will (or its development). Unless you introduce invented concepts such as "soul" or "god", etc., which means that you quit the rational discussion altogether.
"And everything else that follows (the "2nd, 3rd, nth... second" of your life) is a function of these 2 initial variables"
That's an opinion sprouting from a view on the world that states free will doesn't exist. It may be true, but goes counter to many people's intuition that there is a free will.
A good argument would explain how, from the assumption of 'no free will', we can arrive at such things as 'property' and 'crime'. Why would societies sprout up in which people get imprisoned for using stuff that they didn't make themselves, or that they didn't get permission to use for from those who made it?
My argument would be that this is just evolution: local optimization of the non-physical world, in the same way evolution gave us all kind of complex physical structures. On the one hand, I can't see much fault in that, but on the other, giving up the idea of free will means reintroducing predestination, accepting that no criminal can be blamed for their crimes, no Isaac Newton or Olympic champion can be praised for their accomplishments, that capital punishment and war are just things that happen, etc, and even accepting that there is no such thing as 'me'.
So, yes, if physics as we know it is correct, logically, I think you are right, but emotionally, I don't believe it. That might have to mean that somewhere in our bodies there is a part that can bend the laws of quantum physics as we know them.
This world view is a very 19th century clockwork universe one: if we know all the initial conditions in sufficient detail, we can predict the outcome. There are two problems with this: at a fundamental level, quantum mechanics says you can't know the initial conditions with infinite precision[0]. Second, even at a classical physics level non-linear systems with very similar starting points can have very different final states. As a practical matter that means even if the rules are deterministic, you cannot predict with 100% accuracy what the outcome will be.
So we have a complex non-linear system (your genes) and a very very complex non-linear system (the environment) that make a very complex non-linear system (your brain). Because we have all of these complex non-linear system interacting together, we cannot determine what will happen in any practical sense. We can probably make some observations that will generally hold about some decisions a human brain will make (e.g. it's been a few hours, it will be hungry and decide to eat), but these are more probabilistic rather than 100% concrete deterministic rules.
Libet's study is terrible -- as with many of these studies it's famous for its controversial conclusion, but its flaws are very obvious.
He specifically asked his subjects to listen for an urge rather than applying free will ("let the urge [to move] appear on its own at any time without any pre-planning or concentration on when to act”). That the urge is measurable before the decision, then, just means his subjects did what he asked them to -- to wait for a feeling of an urge.
And frankly, it's most likely the urge was the tension of having sat there a while, knowing someone's expecting you to feel an urge to move your arm...
1. Ask subjects to wait for an urge to react to (and therefore not apply free will)
2. (Nice-sounding but actually irrelevant measurement set-up with dots, clocks and electrodes)
3. Observe subjects decided after a measurable urge
4. Conclude free will doesn't exist, rather than that your subjects did what you asked them to in step 1.
Yes, indeed it is a simple flaw. One of my criticisms of the paper, given how popular it is to write about it in the media, is quite how obvious the flaws are.
>Also the strict separation of conscious/unconsciousness seems simplistic. Most actions may very well be processed subconsciously with very little decision making bubbling up to consciousness.
This. The "unconscious" processing an answer/reaction belongs to the same entity that the conscious part belongs too.
Implying we're some kind of automata because reactions can come pre-processed by the subconscious part misses the point.
The subconscious part is the same "self" that the conscious part is, it's not just what we can think "out loud" in our head that counts as us taking a decision.
This is scientific evidence, not mathematical proof.
One way to frame the concept of free will is that some subjective experiences cause rather than are caused by or are identical with, physical changes in the brain.
One way to test for causation is to see which one happens first. If free will (at least as defined above) is true, we expect that some subjective experiences should precede the corresponding physical change in the brain. If free will is false, then we should expect subjective experiences to be simultaneous with or come after the corresponding brain change.
Since the actual result of the experiment (brain change comes before subjective experience) is less likely under the "free will" hypothesis than it is under the alternative, by Bayes' rule, it should make you less confident of the existence of free will. If you believe that this result would be only slightly less likely even if free will were true (as you apparently do), then you should update your estimate only a little bit, but this is still the right kind of experiment to run if you want to test for free will.
>One way to test for causation is to see which one happens first. If free will (at least as defined above) is true, we expect that some subjective experiences should precede the corresponding physical change in the brain. If free will is false, then we should expect subjective experiences to be simultaneous with or come after the corresponding brain change.
Only if you assume that "free will" implies only conscious decisions.
Your subconscious brain is part of the same self as "you". So whatever you think as your "free will" is also part of that, just that part of it plays as "voice inside your brain" (conscious thoughts) and some of it does not.
It's not like the subscoscious processing is done by some other entity and imposed upon you.
In my opinion it is almost trivial to see that something is fishy with the idea of free will - just try to formalize it.
Assume you are sitting in a restaurant facing the decision to either order a steak or some pasta. One possibility is that your decision is a deterministic choice, a function of the current state of your body (What nutrients do you need?), your brain (What are your past experiences with steak and pasta?) and your environment (What does the steak on the neighboring table look like?). This does not resemble what I would call free will.
The other extreme is that your decision is completely random and not influenced by anything. This assumes that there is real randomness in our universe which is, as far as I know, an open question. There is no real randomness in classical mechanics, only apparent randomness due to ignorance of microscopic degrees of freedom. Quantum mechanics seems to have probabilistic features but they are, as far as I can tell, at odds with the unitary evolution of quantum systems and it remains to be seen whether there is real randomness in quantum mechanics or not.
But lets just assume there is real randomness, at worst, if there is only apparent randomness, this option becomes deterministic and degenerates into the first option. There is still some freedom in this option, namely the probability distribution over the different choices. This probability distribution may just be what it is for no deeper reason, a fundamental property of the source of randomness. In this case I wouldn't call it free will, too, because the choice is entirely random.
It may also be the case that the probability distribution gets shaped in a deterministic way. The steak on the neighboring table looks really good making it more likely that you choose it but there is still some probability that you will choose pasta. This is kind of a middle ground between the first two options, the final choice is random but the probabilities reflect your current and past states and the state of the environment in a deterministic way. But again I would not call this free will.
So what would free will have to look like? The choice must not entirely depend on the current state of you and your environment but it must also not be completely independent of it, i.e. be completely random. I spent quite some time thinking about this but I am completely unable to come up with something that is in some sense between deterministic and random (including deterministically shaped randomness). Am I - or even everyone - missing a (fundamental) third option? Is thinking about free will in terms of systems and states and state changes in some way inappropriate? For the moment I will side with the people denying the existence of free will, if someone can formalize what free will really means I will reconsider things.
The law of causality is ultimately incompatible with free will. You cannot have truly autonomous agency within a physical framework ruled by cause and effect. To truly express freedom of will, without influence from any prior cause, you would literally need to exist in your own isolated vacuum of space and time (no outside influence from environment), having also created yourself (no outside influence from genetics). That is to say, if you believe you have free will, the burden is on you to prove that your body, and your brain itself, are not bound by the laws of the physically causal universe.
reply