I almost never listen to podcasts or podcast-style videos simply because I can read faster than I can listen to people talk, and because podcasts are invariably about 1.5 hours long, which is TL:DL for me. But I didn’t listen to nearly all of the following video (the beginning isn’t relevant) as Michael Shermer sent it to me touting it as a pretty convincing argument for free will. Yes, it’s an hour and a half long, but you can skip the first hour and still get the meat of the argument.
The arguer is Dr. Christian List, a professor of political science and philosophy at the London School of Economics and Political Science. He’s just written a new book, Why Free Will is Real, and that’s what List and Shermer talk about in the podcast (click on screenshot below) and in the identical video (below that).
Here’s the description of the video, a descrioption which is pretty accurate. I’ll put a few comments below the video.
Philosophers have argued about the nature and the very existence of free will for centuries. Today, many scientists and scientifically minded commentators are skeptical that it exists, especially when it is understood to require the ability to choose between alternative possibilities. If the laws of physics govern everything that happens, they argue, then how can our choices be free? Believers in free will must be misled by habit, sentiment, or religious doctrine. Why Free Will is Real defies scientific orthodoxy and presents a bold new defense of free will in the same naturalistic terms that are usually deployed against it.
Unlike those who defend free will by giving up the idea that it requires alternative possibilities to choose from, Christian List retains this idea as central, resisting the tendency to defend free will by watering it down. He concedes that free will and its prerequisites—intentional agency, alternative possibilities, and causal control over our actions—cannot be found among the fundamental physical features of the natural world. But, he argues, that’s not where we should be looking. Free will is a “higher-level” phenomenon found at the level of psychology. It is like other phenomena that emerge from physical processes but are autonomous from them and not best understood in fundamental physical terms—like an ecosystem or the economy. When we discover it in its proper context, acknowledging that free will is real is not just scientifically respectable; it is indispensable for explaining our world.
I’m not going to pull any punches here: I think List’s argument is weak, bogus, and so opaque that it’s not clear what the sweating professor is trying to say. His argument is for “free-will emergentism”, which is not a form of compatibilism, but a real claim for you-could-have-done otherwise libertarian free will in humans. He’s not making the argument of Dan Dennett or Sean Carroll that, although our behaviors are determined by the laws of physics, there is a way in which we can sensibly speak of free will. Dennett and Carroll are determinists who are compatibilists, seeing that some conception of free will can still be useful in human society. As I’ve said, I see their arguments as largely semantic, and feel that the truly important thing to understand is determinism, for determinism has profound social implications while, at best, compatibilism lets use use the term “free will” and, according to some philosophers, including Dennett, provides an essential form of social glue without which civilization would fall apart.
List gives three criteria for true free will: intentional agency; causal control over one’s actions, so that you have alternative possibilities to choose from and could have chosen otherwise; and there are higher-level aspects of human behavior, including “choice”, that are not reducible to physics and chemistry. In other word, some form of true libertarian free will arises mysteriously between the molecules that make up our brain and the behaviors that emanate from that brain.
Sadly, I cannot find anywhere in List’s spiel where he say how this emergent free will arises, or how it manages to defy the laws of physics. He uses weak analogies, like saying that while the “weather” arises from motions of molecules in the atmosphere, meterologists use models that abstract from the microphysical to the macrophysical and are indeterministic, giving probabilities of weather events. But that’s a bogus analogy, for the macro-“weather” is certainly consistent with, and arises from, lower-level phenomena. True, “rain” is an emergent property, but it is absolutely consistent with the laws of physics. List’s free will isn’t. Were I to be uncharitable I would say that List’s “top-down” free will arises from the effects of a tinfoil hat.
Remember again: List is not a compatibilist but a true libertarian who accepts the kind of free will limned by the Abrahamic religions. His failure to explain the source of that free will is the big flaw in his argument, which, to my mind, isn’t worth dissecting further. As Hitchens’s Razor states, “What can be asserted without evidence can be dismissed without evidence.”
List has no evidence, just an assertion that some people find psychologically soothing.
sub
I haven’t listened to any of the podcast yet, but this
Emergent behaviour is not autonomous from the physical processes on which it is based. The fact that humans find it hard to analyse the behaviour of an emergent system in terms of the behaviour of its components does not mean the emergent behaviour is not constrained by the behaviour of the components.
If the behaviour of each of my neurones is deterministic, then the overall state of my brain (i.e. me and my thoughts) must also be deterministic.
Oops. Forgot to complete the first sentence. It should end “is a pile of dingoes kidneys”
List is not saying that emergent phenomenon don’t follow the rules of physics. I think our host gets this wrong from the podcast. List is saying that it is matter of differing levels of description. Free will is part of how we think about human agency. Talking about how we don’t have free will because of physics is about as meaningful as saying that our morals must obey the rules of physics. I suppose they do because everything does but it’s just not a meaningful or useful observation.
Causation is a term that has no meaning at the lowest level of description of which we are aware–quantum mechanics. This is not controversial (although it is weird)
Causation is an emergent phenomenon that arises, stochastically (and incredibly reliably) when you get enough random events happening together (sort of like how I can roll 1000 dice and know that I will not get 1000 6s).
I don’t see why freewill can’t be just as emergent as causation itself? Dennett (and Hume before him, and most philosophers really) argue that its just ‘being caused in the right way’.
If someone wants more than this (e.g. at 1 billionth of a nanosecond before a decision it could go either way) then this is something they cant have. But its something that they shouldnt want anyhow, becuase it neurolgically naive and silly.
Most of us want our beliefs to be causally connected to evidence, our decisions to be causally connected to reasons, and our morals to be causally conencted to principles. I really dont feel the difficulty.
The impression I get is that most people do not want their decision to be causally connect to reason. They want full metaphysical libertarian to be real and they have a real difficulty accepted that the universe is deterministic.
Quantum phenomena are causal. EPR, Cat, or NOON states are intrinsically linked to causal events prior to their existence. This is how they are empirically generated, studied, and interpreted.
Quantum phenomena are causal, depending on your definition of “causal”. If you insist on time-asymmetry in your “causality” (where a cause determines an effect but not vice-versa), then only many-body phenomena, where entropy is a meaningful quantity and irreversibility applies, are causal.
IMO, the problem is that “Different levels of description” can’t get you to “there is something more going on than just molecular composition, properties, and motion”.
Different levels of description can be very accurate, and very useful. But they don’t in themselves create or justify a claim of violations of physics.
No violations of physics. List is not claiming that.
He’s implying it. Else how do alternative possibilities emerge from a system that has none?
And I’m guessing he has no testable biological hypothesis about that how, does he? It’s handwaving.
He’s not implying it. His whole explanation acknowledges determinism though his definition of free will is independent of that.
I haven’t heard the podcast yet, but I’ve read some of List’s papers. You nailed it.
Yes, that would seem to be a no-brainer, but SO many people can’t (or won’t) grasp it.
Yes exactly. Put another way, ecosystems and economies still obey the laws of physics…and so do brains and minds.
Minds, ecologies, and economies may be similar in that we cannot predict how they will behave with our limited understanding of all the forces at work in them (and how to accurately model those forces). But AFAIK no serious scientist or philosopher is running around claiming there are emergent phenomena granting ecologies a freedom of choice not available to them via the laws of physics.
I haven’t finished the podcast yet but, based on the first half, I completely agree with List. I respectfully submit that you misunderstood his argument. Admittedly, he didn’t really present it that well. That may be due to the Q&A format. I want to get his book, though it targets professional philosophers which I am not. I like Shermer but he didn’t really ask too many probing questions. Perhaps it was because he was on List’s side, which he freely admitted at the start.
List is not suggesting that the weather doesn’t arise from lower-level phenomenon. Instead, it is at a different level of description. I think this is also Sean Carroll’s argument though I am not 100% positive. I’m hoping Sean reads List’s book or listens to this podcast and comments.
I don’t really understand what you mean by “source of free will”. It’s the term we use to describe human agency. I suppose List’s three criteria could be considered the source but I don’t see that as particularly helpful.
List used weather analysis as an example of where scientists who, presumably, believe in low-level physics and determinism, still choose a higher level of description involving objects and forces which they do not derive from the low-level description. They acknowledge that they could but only in principle. They also believe that, even if they could derive weather from low-level knowledge of particles and forces, it would not be useful.
I believe the major source of misunderstanding here is that List’s free will is the result of how humans think of the world. In a sense it’s a choice, much like the choice meteorologists make in how to describe weather phenomenon. The determinists derive lack of free will from physics. Perhaps if List was able to present it more completely and the ideas in an order of his choosing, it would result in a more convincing presentation. I look forward to reading his book.
I respectfully submit that you don’t understand List’s argument. He adumbrates a you-could-have-done otherwise source of free will. It is manifestly NOT compatibilism where at a “higher level” you can get true libertarian free will. He explicitly says so.
Yes, the weather is at a different level of description, but he uses it (AS IS SAID) as a false metaphor for his completely different argument: that libertarian free will arises as a non-deterministic phenomenon. Note when he says that his kind of free will is a “you could have done otherwise” free will. That shows it’s libertarian and not compatibilist, and against the laws of physics. It’s mystical.
Sorry, but listen to the whole podcast. It is NOT Sean Carroll’s argument, and it is not compatibilism.
I have now listened to the whole podcast and even revisited bits to make sure I didn’t get the wrong impression. I haven’t changed my position, which is probably not a surprise.
I don’t see how you get the idea that List doesn’t believe in determinism. Instead, he is saying that it is irrelevant to the level of description to which the free will concept belongs. His explanation of this and his weather example start at around 39:30.
I believe you and Sam Harris are somewhat motivated in your views by the potential for reform of our criminal justice system. As List says at 1:24:00. “You don’t have to be a free will skeptic in order to be critical of certain crude and outdated approaches to punishment.” This is what I’ve been saying. There are very good reasons for reform that do not involve free will at all.
Yes I struggle to see how this debate is particularly helpful in determining the need for criminal justice reform or the direction that reform should take. I think that the kind of reforms that I believe Professor Coyne would like to see made to the justice system (which I would also like to see) can be argued for without reference to the existence or otherwise of free will.
I also wonder where exactly the rupture is in the chain of determinism that dictates that a felon could not have done otherwise but we (as a society) can choose the type of justice that we mete out to that felon.
It seems to me that for practical every day purposes such as reforming the criminal justice system, voting in elections, carrying out our jobs, choosing what to have for dinner or deciding what to comment on a thread beneath a web-site post we all behave as if we have free will and expect others to also have free will. We may know that fundamentally our actions and choices are determined by chains of events dictated by the laws of physics and stretching back to the big bang but that is not information that we can make practical use of in our day to day lives. The only way we can manage our lives in any sort of sensible manner is by tacitly accepting/pretending that we have free and can choose between alternative actions even if we acknowledge that deep down this is an illusion.
“I also wonder where exactly the rupture is in the chain of determinism that dictates that a felon could not have done otherwise but we (as a society) can choose the type of justice that we mete out to that felon.”
This and similar arguments have been made many times and I have yet to see any kind of response from incompatibilists. We either are all robots who don’t make choices (and think we do) or we aren’t.
I don’t know what kind of response you are looking for, but I think Jerry covers that quite well when he talks about the deterant effect of the law.
It seems pretty simple to me. We are determined by stimuli from our environment. Well everyone is a constituent of everyone else’s environment. Therefore speaking up for what you believe is a stimulus for other people to consider adopting your beliefs. Including that “we” (our past selves and actions taken) are constituents of our own current environment. (For example, we put a reminder on the fridge to stay on our diet, or whatever.) No Free will needed. Just actions that shape the environment that in turn stimulate us be who we become.
I don’t see how that explains anything. If the criminal didn’t really have a choice, neither do we in giving out punishment. If the criminal is not morally responsible for the crime, I’m not morally responsible for sentencing him to death. That it was the result of the environment and/or someone’s argument is all determined. There are no choices being made, just physics doing its thing.
Jamie,
but I think Jerry covers that quite well when he talks about the deterant effect of the law.
That doesn’t address the problem.
The very fact that we sentence anyone at all, that it can be justified, only makes sense if the person “could have done otherwise.”
How would it make sense to even set up the “deterrent” of sending anyone to prison in the first place if we don’t distinguish between situations in which someone:
1. “Could not have done otherwise.”
vs
2. “Could have done otherwise – yet still chose to commit the crime”
We need to be able to distinguish between the 1st and 2nd scenario in order to justify sending anyone to prison.
Vaal,
“The very fact that we sentence anyone at all, that it can be justified, only makes sense if the person “could have done otherwise.””
No. There are lots of reasons (and justifications) for sanctioning and restricting the liberty of someone that have nothing at all to do with whether they are morally responsible “free agents” or not. To prevent further crime. To protect the innocent etc. etc. These are directed toward the present and the future, not the past (he could have done otherwise).
Jamie,
I made no mention of “morally responsible “free agents.”
You give as one reason to put someone in prison:
To protect the innocent etc. etc. These are directed toward the present and the future, not the past (he could have done otherwise).
Again…the reason you would determine that someone needs to go to prison to “protect the innocent” is that the criminal “had the capability to do have done otherwise – but chose not to.” It is the conjunction of those two propositions that tells you what you need to know about what type of person you are dealing with, to assess whether we need to worry about him enough to send him to prison.
And the only possible justification for rehabilitative system is that “someone COULD do otherwise.”
But this particular thread I think is getting a bit long. If you’d like to challenge this point, perhaps you could do so in reply to my post I made further down the page. Cheers.
‘I believe the major source of misunderstanding here is that List’s free will is the result of how humans think of the world.’
That is not new, is it? That view is the naive model of free will. It is how we model human beings. It does not imply the impossibility of predicting behaviour.
‘They also believe that, even if they could derive weather from low-level knowledge of particles and forces, it would not be useful.’
Even if we could only make the same prediction from a microscopic model that we make from the macroscopic one, it would still be as useful, although it would not be as efficient.
It seems to me that List is saying that the the Intentional Agent model is a good one. So we must accord it a place in reality. Thinking of matter as continuous is a good approximation at some scale. However, we know it to be untrue.
He also says that a system described by the Intentional Agent model is deterministic — modulo QM, I assume — at a microscopic level. So we would, in principle, be able to compute the macroscopic behaviour, thus, through much labour, removing the need for his Intentional Agent model.
It doesn’t remove the need for the Intentional Agent model. Human beings use that model for evaluation, not just for prediction. Should I choose a career in law? Knowing the future position of a whole lot of atoms would not answer the “what career should I have” question – I have to translate those atomic motions back in to a human-level description. Which will be an Intentional Agent description.
I agree. It does not remove the need — in fact, seen in retrospect, that is not what I meant to say. The IA model would still be useful.
Isn’t his analysis of the courtroom scenario flawed as well? If the defendant’s behaviour was determinable, then we have to face that fact. It may or may not be relevant to the judge’s decision — that is a different matter.
He also talks about the difference between an obviously impaired person and a healthy one. He says that the no-free-will theory does not draw a distinction between them. That is true, but also irrelevant.
He talks about two models of the same system. However, don’t the models have to be consistent with each other? He seems to admit that the system is determinable at a microscopic scale, but he does not state why we would lose determinism at any stage.
So his idea is that the IA model is a reasonable one at this time. However, we already know that.
This part of List’s presentation I have some trouble with. I don;t think he’s wrong on the details but it confuses his audience. He talks about systems being determinate at the micro level but indeterminate at the macro level. Meteorologists do treat the system they are studying as indetermined. They certainly don’t derive their predictions from micro-level particle positions and forces. However, this is not because they deny determinism but because it is the only useful choice available to them. Unfortunately, describing it this way leads some to conclude that List is pulling some sort of trick or involving woo. He is not.
I’ve never understood the argument about free will. If we live in a deterministic universe, then given the exact status of the beginning of the universe and a powerful enough computer, I can calculate what you will have for lunch today.
The brain is an electro-chemical machine. It has free will in the sense that it CAN decide whatever it wants to decide, but given sufficient understanding and computing power, what that brain WILL decide can be calculated.
The only really interesting thing there is that the brain not only takes external data, but can feed data back into itself.
The one monkey wrench is the possibility that physics is not deterministic at the quantum level. I’ve heard people assert that, but I don’t know if it’s true or to what extent, if true. (This may be yet another case of people misunderstanding the observer effect)
Yes, and so it seems from the Bell inequality experiment. But randomness still doesn’t give us agency, so even if there is pure quantum indeterminacy (which, by the way, somewhat attenuates our ability to predict), that doesn’t mean we could have CHOSEN otherwise. If agency is anything, it is that we don’t make alternate “choices” based on the indeterministic movement of particles.
I suppose on a certain level, all of choices are based on the movement of particles. 🙂 Of course, that doesn’t give us a mystical soul of free choices. As I understand it, it’s more likely just to be a “random factor” that prevents me from calculating what you will have for lunch today with certainty.
Amusingly, the way theocrats square the circle of free will with an omnipotent god has utility. They say “You can decide what you want, but god knows what you WILL choose” or something to that effect.
Instead, you can turn it around to “your brain has the potential to make any decision, but physics can calculate what decision it WILL make”. So your brain has the potential to decide to run through the streets naked, but I calculate that you won’t.
The brain is incapable of all decisions. In fact, it can be shown the brain is incapable of calculating the trajectory of all the sand blown in the Mojave desert on a summer’s afternoon. Much like the body is physically not capable of swimming 1500m in 1s.
Physics restricts not only the limits of our decisions but all our physical processes.
With regard to prediction, it may be the case that precise predictions are impossible in our universe given finite energy (information) and time to modeling future states.
It is either ontologically random or nonlocal or retrocausal.
List does not disagree with determinism. He is a compatibilist but is saying that free will exists at a level of description way above that of fundamental physics, the level at which determinism matters. He is also not really a libertarian or, if he is, he’s one that accepts determinism.
Okay, you needn’t repeat your take on List again. I disagree with it and do not see him as a compatibilist. He SAYS he’s a compatibilist but he accepts libertarian free will at the behavioral level.
I feel unqualified to weigh in seriously on this general matter, both too stupid and too non-industrious over the years to have a strong opinion.
But I’ve had a few queries in my head when this topic arises, though below has nothing specific to do with the video here.
The main query is whether it really is entirely a matter of classical physics, or whether e.g. some of the ideas of Roger Penrose in general, and with Stuart Hameroff claiming quantum effects within the brain, might eventually be accepted. (They are not now accepted by most scientists I think.) So I have attached the abstract of a Hameroff article from 2012 below which is certainly right on the topic.
But before leaving that with you, I should add that my leaning towards the Everett ‘interpretation’ of quantum theory implies I’d be inclined to disagree with Penrose at a more fundamental level, but related here.
And so I again got out that important David Wallace book (same 2012!): “The Emergent Multiverse”. In the page 164 footnote: “If this is a problem, it is not specific to Everett: It is the ancient debate of free will vs. determinism. Again see Box 4.1…” The latter goes for about 4/3 pages, 135-6, the latter including “.. a clash between free will and determinism but better understood as a clash between free will and mechanism.” I take this to mean that quantum theory is not uninvolved in the debate, independently of Penrose-type stuff. Just below that, Wallace is willing to proffer his own ‘feelings’: “For the record, I don’t think there is any incompatibility between freedom and mechanism..” referring to Dennett’s 2003 article and an Oxford 2002 book “Free Will” edited by an R. Kane with many positions on this debate. In any case, Wallace seems to feel that the Everett stuff, though inducing some novel scenarios (mainly the splitting of the observer into different ‘worlds’ where she ostensibly makes different decisions) does not really do any more to advance the problem than do the other interpretations.
Here’s the abstract:
How quantum brain biology can rescue conscious free will
Stuart Hameroff
Abstract
Conscious “free will” is problematic because (1) brain mechanisms causing consciousness are unknown, (2) measurable brain activity correlating with conscious perception apparently occurs too late for real-time conscious response, consciousness thus being considered “epiphenomenal illusion,” and (3) determinism, i.e., our actions and the world around us seem algorithmic and inevitable. The Penrose–Hameroff theory of “orchestrated objective reduction (Orch OR)” identifies discrete conscious moments with quantum computations in microtubules inside brain neurons, e.g., 40/s in concert with gamma synchrony EEG. Microtubules organize neuronal interiors and regulate synapses. In Orch OR, microtubule quantum computations occur in integration phases in dendrites and cell bodies of integrate-and-fire brain neurons connected and synchronized by gap junctions, allowing entanglement of microtubules among many neurons. Quantum computations in entangled microtubules terminate by Penrose “objective reduction (OR),” a proposal for quantum state reduction and conscious moments linked to fundamental spacetime geometry. Each OR reduction selects microtubule states which can trigger axonal firings, and control behavior. The quantum computations are “orchestrated” by synaptic inputs and memory (thus “Orch OR”). If correct, Orch OR can account for conscious causal agency, resolving problem 1. Regarding problem 2, Orch OR can cause temporal non-locality, sending quantum information backward in classical time, enabling conscious control of behavior. Three lines of evidence for brain backward time effects are presented. Regarding problem 3, Penrose OR (and Orch OR) invokes non-computable influences from information embedded in spacetime geometry, potentially avoiding algorithmic determinism. In summary, Orch OR can account for real-time conscious causal agency, avoiding the need for consciousness to be seen as epiphenomenal illusion. Orch OR can rescue conscious free will.
It has been pointed out that quantum theory obviously is responsible for the consciousness, to the extent that it describes reality of course. Penrose’s argument is really only about where in the neuron quantum mechanics comes into play and to what extent it matters.
The real problem with Penrose is his motivation. He claims that a mathematician’s wondrous ability to see the truth in a bit of mathematics is something that requires quantum mechanical level effects. To me this is absolute woo. His theory is just the result of taking two things we don’t yet understand completely, the workings of the human brain and the theory of everything, and thinking they must have some common cause. It’s junk science!
Are you aware of Penrose’s many accomplishments?–in mathematics itself, the tiling stuff, the twistor stuff? –in theoretical physics (via mathematics of course) the first deduction from General Relativity, with a few very believable extra assumptions about causality and energy, of the absolutely certain existence of black holes (well before there was empirical evidence)?
He may be mistaken in this matter and in the related question of whether the human brain can operate beyond the level of Turing computability, but it’s not junk.
Junk appears when people with superficial knowledge, and little realization of that fact, start blurting from a keyboard only weakly connected to a brain.
Yes, I most definitely am aware of Penrose’s accomplishments. I am only saying this stuff is junk science and that’s not just my opinion. They’ve gotten a lot of blowback on this from real scientists. As usual, a good place to start is:
https://www.wikiwand.com/en/Roger_Penrose#/Physics_and_consciousness
The word “junk” appears nowhere there. Perhaps you could point to a synonym.
“Junk” is not in that wikiwand. Is a synonym?
The “junk” part was mine. Sure, it is somewhat lazy on my part but why should I write my own refutation of his theory when there are so many already written? I stand by “junk” as it is completely sloppy thinking, IMHO. People seem to be calling out my use of this term as if I don’t have the right to question the work of Sir Roger Penrose. Maybe, but I do it anyway.
I called Penrose on his nonsense about AI as (what one would call in the US) an undergraduate junior, so whatever …:)
The physics stuff I also knew was crap, but didn’t deal with – I was working on philosophical opposition to AI, so went broad rather than deep on one.
Met him the following year, by chance, and asked him Dennett (and also Feferman’s) question from _Darwin’s Dangerous Idea_. He had no idea that there was a disciplinary difference on “algorithm”.
As for the physics of the brain, Hameroff and a bunch of those guys have been harping for decades and don’t seem to ever pay attention to Grush/Churchland or Vic Stenger’s criticisms (including, in the latter case, calculations of decoherence time).
Another reply to Paul’s above, but also remarks on Keith’s further down:
To be fair, and as Paul likely knows, there are many good scientists who spent effort refuting (or attempting that) Penrose’s two books to do with AI and Godel’s theorem. So there are much stronger voices than what he suggests favouring Paul’s viewpoint here, if not his intemperate use of “junk”. Paul’s URL really only gives Minsky’s offhand brief derisive comments. Looking at the approximately 70 year history of AI, despite his prominence, it has turned out that Marvin Minsky’s work has been shown to be almost universally mistaken, with some easily found zinger’s about the upcoming (never happened) AI successes about 15 or 20 years into that history. So I don’t have strong opinions about “strong AI cannot work” either, similarly to my indicated earlier ‘sentiments’ disagreeing with Penrose’s more general position referred to in that Hameroff abstract.
The more convincing arguments against Penrose vaguely referred to above are from serious scientists who do not waste their time arguing against stuff that one would derisively refer to as “junk science”; they do and should take Penrose seriously, as a few people here apparently do not.
So that also applies to Keith, even though I do not have a strong enough opinion to either agree or disagree with him. But I’d be interested to know just what ‘physics stuff’ Keith was talking about when he says “The physics stuff I also knew was crap..”. I’d hope it was ‘only’ Penrose’s ‘state vector reduction’ proposals, and not all the other stuff from 1960 onwards, black holes, twistor theory (e.g. Ed Witten and Michael Atiyah take and took that as very important).
In any case, I hope people reading this non-blog are aware that there are voices of considerable weight, not only Daniel Dannett from the philosophy side, who do not have exactly the same opinions voiced by a majority here.
Of course my link to the Wikipedia page is just a starting point for someone investigating the controversy over Penrose’s claims. This is why these pages have references.
Based on the reaction here, I regret calling it “junk science”. It was shorthand intended to reflect my opinion that, while Penrose has done brilliant work in other areas, this theory of his is not. This is my opinion, offered for whatever it’s worth, and that of many other scientists. It is possible that Penrose is right and we’re all wrong. That’s the way science works.
Yes, I mean only the physics-of-the-brain stuff he proposed when I mention physics here.
Whether or not strong AI is possible or not, the argument in Penrose’s books on the subject is bad. In addition to the misunderstanding of the incompleteness theorems and a failure to grasp what an algorithm is in the CS sense (despite his citing Turing about fallible machines), there is also the smell test – read the bibliographies: what papers or references in AI does Penrose actually cite? This in itself is suspicious.
Libertarian free will will never be shown to be true since it is incoherent/logically impossible. Hameroff’s efforts are therefore futile. You can still choose to believe in libertarian free will if you want, but that is based on faith/feeling alone.
Maybe one could imagine that a supreme being, like a super-AI, could view logic differently and thus make sense of libertarian free will. But I doubt it. Elementary logic, like the law of non-contradiction, is not like solving a super complex math problem.
I’m sure both Hameroff and Penrose are well aware of the many subtleties concerning libertarian and alternative versions of free will and of Dennett’s thoughts. So this seems not really a cogent response to anything above by me, but perhaps you can elaborate.
I have stated no strong opinion of my own above because, as I said, I’ve not done the work, perhaps not even capable of doing it, in order to form such an opinion. But there is a shallowness, not to Jerry’s thoughts, but to much of the response, which I hope might be improved by some serious reading including Wallace’s book, and some of Penrose’s, not just rewording of reviews by science popularizers.
I responded to Hameroff’s claims rather than you personally and just wanted to state how extremely unlikely I find it that the libertarian position will ever be made to make sense. Penrose doesn’t believe in libertarian free will as far as I’ve understood and has left Hameroff to pursue it.
My statement stands, since any conception of free will that is not compatible with determinism is incoherent, no explanation for how it is supposed to work can ever be given. Hameroff has also stated that his retrocausal theory could be interpreted as deterministic.
Daniel Dennett sometimes quotes a libertarian (Randolph Clarke?) that has admitted as much that a “miracle” must take place when we make a choice. This is why most of the scholars that argue for libertarian free will have a religious bent.
The complexity of certain emergent properties, such as the weather, makes them unpredictable; it doesn’t make them contra-causal.
Does the good perfesser think a cheetah exhibits “free will” when it chooses which gazelle to cull from the herd for dinner?
Again, List does not deny determinism and that, ultimately, the weather and everything else is determined. He’s simply saying that determinism is not useful at the level at which meteorology is done. Same for human agents and their free will. It’s not dispute over the nature of reality but how humans choose to describe their choices and actions.
Okay, you are dominating this thread and doing so by characterizing List’s views (in a way I disagree with) the same way in every post. That’s enough. He is NOT a compatibilist in the usual sense; otherwise he wouldn’t have written a book with his “new” theory, which is not new if he’s just saying what Sean Carroll said.
Listening to the podcast I was amused to learn that Shermer also has a dog named Hitch, named after Christopher Hitchens. I wonder if he got the idea from me.
Regarding the denial of free will due to physical determinism, I have several objections:
1. Even if the argument is valid, it has zero explanatory power. We will never be able to model and predict the behavior of complex emergent systems from physical first principles, and it’s impossible even in principle because quantum mechanics is indeterminate.
2. People have a vivid *feeling* of free will, and we will continue to have it forever — even people who accept the determinism argument.
3. Physical determinism appears to me to be incapable of solving the hard problem of consciousness as defined by Chalmers. This suggests (to me) that there are phenomena — namely qualia – that exist independently from and unexplained by physical first principles as we currently understand them.
4. It is premature and unjustified to assume that we understand physical first principles in every detail. We don’t.
Agreed though perhaps I see the consciousness issue a bit differently. I believe we will eventually know how the brain works at the neuronal level and, perhaps, at the level of fundamental physics. This will include perception and consciousness. However, a description of qualia at that level will just not be useful to, say, a musician, an artist, or a psychologist. It won’t be wrong, of course, just irrelevant.
When I have been under anesthesia, I have not had any feeling of free will. Indeed, there was “nothing”, no time, no world, no existence. That tells me consciousness is a part of the brain..no mystical qualia needed.
Obviously one can be unconscious. The problems arise when one *is* conscious.
My point is that we already know how to turn it off and on. I believe as we further study the brain, we will learn how it evolved and no new physics will be involved. While it may be a *hard* problem, I don’t think it is impossible problem.
Shermer recently wrote in a Skeptics article that humans could not even fathom nothing rather something. I just didn’t agree with that with my experiences with surgery. Nothing is just being unconscious.
If you could measure the quantum state of every particle of every atom of every molecule of every neuron in a brain, which you can’t do in practice or in principle, you would still fall short — far short — of explaining the experience of seeing the color brown.
How do you know that?
Because I’m looking at a brown thing right now.
I hope you don’t actually think that is an adequate explanation.
I should have offered a more robust challenge. Perhaps by referencing Dan Dennett’s discussions of qualia from Consciousness Explained or something. But my main point is your assertion that complete knowledge can’t “explain” brown isn’t much more than an unsupported claim. Maybe you’re right, but only “maybe”.
The obligation is on you to explain how a microscopic physical description of a brain state could explain the subjective qualia of “seeing brown”. It’s not on me.
Even if we knew in detail why someone experienced brown the way they do, it would still not be equivalent to that person’s actual experience of seeing brown.
That’s the point.
I leave the argument for Dennett who is far more eloquent than I on the subject. But you are the one who asserts that there can be no explanation. And that, I think, is on you to support with more that an assertion.
My dad was a painter. It was a fun experience watching him mix primary colors to get brown.😀
He could do no other.
Perhaps not only could I have done otherwise but I did do otherwise and the universe split into many worlds and the consequences of that other choice are no longer available to me.
Are you THAT sure about the “many worlds hypothesis”?
The multiverse works for everything, down to electrons.
On the other hand the multiverse is presently unverifiable, so does not buy any space proving anything about free will.
“..the multiverse is presently unverifiable …”
The existence of quasi-classical splittings, and so the multiverse, is a logical consequence of just taking seriously bare quantum mechanics (or even its relativistic version in quantum field theory). And by not accepting that there is a so-called measurement problem requiring the fig-leaf Copenhagen collapse philosophy, or similar in the deBroglie-Bohm “multiverse in a state of denial” theory. Get that book by David Wallace, or read the papers by him and by David Deutsch.
As for specific ‘verifications’:
Try to explain how a quantum computer, by Google say, would be able to do 2 to the power 72 computations simultaneously (more than the number of electrons on our earth by about a factor of a billion trillion–not including all the other ‘earths’ in the multiverse of course).
Or more down to these earths of ours and our multiverse doppelgangers (Google’s cards are staying close to their vest), explain the Mach-Zehnder experiment with half-silvered mirrors. Both of Deutsch’s famous books make clear the need for the splittings here IIRC.
Why? The classical models of computational complexity theory make assumptions that are basically classical. By saying the “other worlds” are needed, one is trying to reintroduce classical ideas where perhaps they aren’t required.
For an example of this at the next level up (computability theory) see W. Sieg’s work on generalized computing devices and their computability. Yes, we know they are all turing equivalent, but not until 2002 or so did we really have a discussion about it – his [and mine] historical work shows that.
For example: a Game of Life computer is not easily (or at all) covered by the Turing machine mode (one has to handwave about parallel processing).
Disclosure: I was an MS student of Sieg’s so my connection even to Turing is not very far – Sieg’s work started with that of Robin Gandy, at G’s request. (Gandy was one of Turing’s only students.)
“.. where perhaps they aren’t required.”, but cogent alternatives seem never to be offered.
Just that the quantum realm is even more beyond our experience than that of classical mechanics! Why should it “obey” classical principles? Why should all computing devices be circumscribed by the Turing conditions?
Another example model, one relevant to the quantum computation discussion, amounts to pins and segments of “noodles”. One can do a tremendous speedup of factoring with such a machine: or so it seems until one figures out the time to cut the segments. Similarly, some folks (my old colleague from UBC, Amit Hagar, at least years ago, for example) point out that these “set up costs” are hidden in discussions of complexity because they are often constant given the fixed model. But if one changes that matters are different – and they may be relevant in the quantum case (preparing the “qbits”).
A comment on “could have done otherwise.” When I say “I could have done otherwise” I mean that there were alternatives available to me that I “could” have chosen but didn’t. That is, there was nothing preventing me from choosing them *except* my own intention or desire, however that is determined. I evaluated these alternatives and rejected them. I do not think it strains language to call this process my free will. Of course the process by which I make choices cannot violate any physical law.
That process is volition, i.e. will.
There’s a tendency in recent decades to use “will” and “free will” interchangeably, but I think there needs to be a push back against it as it badly muddies the waters.
“Free will” should be reserved for the libertarian stance, compatibilist “free will” is not meaningfully distinct from generically defined “will” so the modifier “free” ought to be dropped as superfluous in the compatibilist case.
Thanks to Prof Ceiling Cat for bringing that podcast to the attention of the class. I’ll have to check it out.
Regarding the consequences of Free Will Skepticism on our judicial system:
Even IF one takes a Hard Incompatibilist stance, and we do away with the phrase “Free Will,” and even if we take deep moral blameworthiness off the table…the Hard Incompatibilist can no more do away with the concept that “people could have done otherwise” than anyone else can. It is absolutely indispensable for making sense of any form of justice, law, and system of consequences including imprisonment.
And the fact that we are actually making decisions to take away anyone’s liberty and force them in to a prison after a crime means it can’t be some “figure of speech” version of “could do otherwise.” It can’t be some “fingers-crossed-behind-our-backs because his/her choice was only an ‘illusion.'” It can’t be “just a habit of thought that isn’t true but we have a hard time shaking it.”
Too much is at stake. Taking serious steps to imprison people on illusory habits of thought or figures of speech would be like the problem Christians have with saying Adam and Eve weren’t real but metaphors: “Then: Did Jesus die for a metaphor?”
We have to be able to say it’s “true,” no fingers crossed behind our back, that a criminal “could have done otherwise” in order to justify subjecting that person to any consequences for their action, or criminal inaction.
Not the magical contra-causal theory of “could have done otherwise.” The compatibilist kind. It’s unavoidable.
To expand on the above, to make it clearer what I mean. Again, put aside the phrase “Free Will” and even put aside “moral responsibility.” What makes the response of censure, or consequences of any time, like being fired, or being sued, or being sent to prison, justified?
Take two different scenarios. In each case, we would have hotel security video documenting each scene:
1. Let’s say “John” is a wheelchair-bound paraplegic on vacation at a resort. His helper has wheeled John in to a position several yards from the resort pool, so John can enjoy the view, and the helper leaves him momentarily. John is the only one at the pool.
Then, a three year old child who has wandered away from the family totters along and falls in to the pool. The child starts obviously drowning. Let’s stipulate John also hasn’t use of his vocal cords so can’t call for help. The child struggles and struggles and then dies right before John’s eyes.
2. After the tragedy, one week later, the resort has hired a lifeguard to watch that pool so this doesn’t happen again. “Sam,” the lifeguard, has passed all the tests required to show he is capable of saving a drowning person. Sam is alone standing on the poolside. Oh no! Another toddler has wandered toward the pool and has fallen in!
The Toddler starts thrashing around drowning. Even looking toward Sam pleadingly.
Sam CLEARLY sees this child drowning only a couple yards away, yet simply stands watching with his hands folded, occasionally just waving “high” to the child until the child drowns to death.
We know that we would rationally treat these two men differently. The first case, John, is a tragedy. Not just for the child, but for John who had to watch the child die. Why wouldn’t we think otherwise and instead censure John for “just sitting in his chair” and not saving the child, or call for him to be fired, or contemplate legal consequences for John? It’s because it was not possible for John to do otherwise even if he wanted to!
But in scenario 2, we would think it justifiable to censure Sam for watching the child die. And to at the very least have him removed from his job as lifeguard (would YOU hire Sam as a lifeguard?), and very likely have Sam face criminal charges for negligence of some sort.
Why? Because unlike John, Sam was capable of doing otherwise than simply watching as the child died. That is, he was fully capable of saving the child, but chose not to.
It is *only* the understanding that Sam “could have done otherwise” that could justify a reason to treat the two men differently in terms of what we think of them, and in terms of having different consequences for Sam vs John.
That’s why the claim “no one could have done otherwise than what they did – if you turned back the clock” isn’t helpful at all in actually distinguishing what we need to distinguish, in the justice system, or pretty much anywhere else. Instead we need the concept of “what someone could have done IF he had wanted to” – in order to determine if this is someone we actually need to worry about, put in prison, or censure.
The thing is the justice system “doesn’t have a choice” too. All we see are patterns of human behaviour. We have no more freedom of choice to fire or imprison someone than he or she has to not mess up at work or commit a crime.
Does a bird have a choice to attack another bird? Does a flock of birds have a choice to fly somewhere?
The fact that human social behaviour is insanely complex doesn’t make us any different.
Andrei
I’m unclear about the point of your post.
As I said, my argument was made in the context of determinism.
Did you mean to agree with, or disagree with my analysis that determining “someone could have done otherwise” is an essential for the justice system?
In other words, the claim “someone could have done otherwise” isn’t unavoidable “because we are determined.” If we take for granted everything we do is determined, simply noting that everything is determined doesn’t settle any questions as to whether we are ever speaking sensibly, rationally or without contradictions.
The proposition “he/she could have done otherwise” is essential for justifying any sort of consequences for someone who has done a criminal act.
Agree?
I agree that there is a distinction. The probability of the action we think is “right” (helping a child etc.) must be non-zero.
Does deterrence make sense in light of no-free-will determinism? I don’t think so.
To deter someone is to influence their future behavior. They could choose A or B. You prefer that they choose A, so you take steps to favor that. You might say, if you choose B you will be tortured to death, but if you choose A you’ll get some number of virgins in the afterlife.
This presupposes they have a choice. It also presupposes that you have a choice.
Making a choice is a deterministic process, not a magical event. Deterrence is thus completely compatible with determinism.
Kosmos,
That’s missing the point.
What sense can it make to “deter” someone from an action if “he could not do otherwise?”
No one bothers to “deter” anyone from something he can’t do. You won’t spend time “deterring” me from turning Manhattan Island upside down this afternoon, because that’s not something I could do. Rather, you deter people from that which it is possible for them to do. And if someone is doing X and you want to deter them from doing X, it has to be possible for them to “do otherwise than doing X” in order to make sense.
To say, for instance, “if you drive drunk we will put you behind bars” can only make sense as a “deterrent” if people could “do otherwise” than drive drunk.
Right?
I took it that Stephen objected to deterrence in a compatibilistic interpretation as well. From previous posts of his I have seen an objection to determinism itself, though I might be mistaken.
Vaal, aren’y you a legendary compatibilist on this site? 🙂 I am a compatiblist and choosing between different courses of action is perfectly compatible with determinism, so no problem for deterrence there.
Kosmos,
Ah, I misunderstood your position.
*does secret compatibilist hand-shake*
Yup I am a compatibilist about Free Will.
But having taken up “legendary” amounts of pixels on the subject in previous discussions, I have to be careful not to overstay my welcome 🙂
I don’t contribute much to free will threads these days, but just thought I’d dive in to this thread a bit as I’d just been debating a similar subject elsewhere and had it on my mind.
Cheers.
Deterrence is one of very many concepts people hold that requires a stance of libertarian free will. Vaal expressed it very well. We all act in day-to-day life as though we have libertarian free will. That’s not to say that we DO have libertarian free will — merely that we think we do, and act as though we do, and we can’t help it. Not to do so would probably get one committed to a mental asylum.
S,
I would differ from what you wrote. I don’t think it takes thinking we have Libertarian free will – of the contra-causal type – in order to get through the day making choices.
I think it takes the freedom we actually have, and generally use in our deliberations: the compatibilist account of free will.
The problem of positing that we need to get through life imagining contra-causal free will still leaves the problem that we would be constantly engaging in untruth and illusion in our deliberations, which actually removes the rational ground of our deliberations. It makes us “just doing determined stuff” vs making sense.
The compatibilist account preserves the coherency of our having “choices” and “freedom” within a deterministic context, so we don’t have to engage in make-believe or delusion.
You have no choice to deter or not. You just do. Same with collectives of people.
My cafe conversation on free will.
Before we became self aware, consciousness if you will, there was no need of free will. Needs of life are instinctive, adaptations, and then you die. After the efforts of evolution via natural selection, emergence of a bigger brain, the arrival of more complex behaviours ensued and eventually out of it , the concept of free will. We were now something special in the universe. Ideologies made sure of this by hammering it home (for centuries) to give individual meaning, control, cultural cohesion, until… science. Not going to blather on about quantum indeterminacy as that is drummed home whenever this subject comes up and i have a limited grasp.
Winding up (or down) we make non free will choices as our existence depends on it and nothing changes this decision making process apart from our understanding of the notion of free will as non free will, as this deepens perhaps, and as the some would like, some decisions will be weighted by the latter.
Coffee is over, so is my lemon cake, is that an omen about my post, i don’t care, the cake was nice?
I think that the best argument that free will as choosing from different possibilities isn’t real is that the possibilities to choose from are not real because they are in the future and the future is not real in the moment of the decision.
Sub
“I almost never listen to podcasts or podcast-style videos simply because I can read faster than I can listen to people talk, and because podcasts are invariably about 1.5 hours long, which is TL:DL for me.”
I tend to agree, and though I started with the audio podcast, I soon switched to the YouTube version, which at least offers the option to play at faster-than-normal (FWIW, it also has the talking heads of the participants). It’s available at https://www.youtube.com/watch?v=9yxHA_gv7vc
“I almost never listen to podcasts or podcast-style videos simply because I can read faster than I can listen to people talk, and because podcasts are invariably about 1.5 hours long, which is TL:DL for me.”
I tend to agree, and though I started with the audio podcast, I soon switched to the YouTube version, which at least offers the option to play at faster-than-normal (FWIW, it also has the talking heads of the participants). It’s available at https://www.youtube.com/watch?v=9yxHA_gv7vc
There’s also a transcript feature, so one can simply read the conversation as text.
I forgot to add that there is a “transcript” feature, so one can read the conversation as text.
I never thought of playing it faster, which is surprising since I know blind people use screen readers that read text to them much faster than normal. I will try it next time I listen to a podcast.
It is also nice to have buttons to skip backwards by, say, 15 secs. Much better than trying to reposition the “thumb” of the video progress bar.
List’s argument leaves out something important, I think. But before I get to that, I want to agree both with one of List’s key points, and one of Jerry’s.
We could have done otherwise, even if determinism is true at the level of particle physics. List is right about that.
Jerry writes:
Yes, exactly! It’s because people don’t sufficiently understand scientific determinism, confusing it with intuitive ideas about causality, that they are incompatibilists.
List’s position is a lot like Sean Carroll’s, in that both of them insist on a sharp distinction between levels of description. Sean Carroll says that you can talk about the level where there are people and actions, or you can talk about the level of electrons and quarks. But you shouldn’t mix them, he says, or you’ll get confusion. List says that human action is indeterministic at the agential level even if it’s deterministic at the particle-physics level.
What gets left out in that approach, though, is that there are useful things to say about how the levels mix. The most important one for the free will debate is that the “arrow of time” disappears at the fundamental-particle level. But our intuitive idea of causality includes an arrow of time. So when we apply our intuitive idea of causality to that level, we get everything wrong.