There’s no scientific reason to believe that we have free will. There’s no buffer zone that we’ve found in any of the physical laws of how the universe works to make room for free will. There’s non-determinism; but there’s not choice. Choice is the introduction of something, dare I say it, supernatural: some influence that isn’t part of the physical interaction, which allows some clusters of matter and energy to decide how they’ll collapse a probabilistic waveform into a particular reality.
Faced with novel situations, humans and most animals spontaneously increase their behavioural variability...
Controlling external events: the input
Thus, competitive success and evolutionary fitness of all ambulatory organisms rely critically on intact behavioral variability as an adaptive brain function. But relative freedom from environmental contingencies is a necessary, but most often not a sufficient criterion for such accomplishments. Tightly connected to the ability to produce variable behavior is the ability to use the effects of these behaviors to control the environment. The incoming stream of sensory information is noisy and fluctuates for any number of reasons. Any covariance between the behavioral variations and those of sensory input indicates that the latter are con-sequences of the behavior and can thus be controlled be the animal. This function is so paramount, that we humans express our delight over control of our envi-ronment (including other people) already as children, by e.g., shrieking in excitement when Daddy jumps after a “boo” or proudly presenting Mom with “look what I can do!”.
Although I should note that I believe there to be phenomena that qualify to be defined as ‘free will’. Specifically endogenous processes generating behavioral variability and thus non-linearity. Especially if you can show that the complexity of transformation by which a system shapes the outside environment, in which it is embedded, does trump the specific effectiveness of the environmental influence on the defined system. In other words, mind over matter. You are able to shape reality more effectively and goal-oriented and thus, in a way, overcome its crude influence it exerts on you. For example, children and some mentally handicapped people are not responsible in same the way as healthy adults. They can not give consent or enter into legally binding contracts. One of the reasons for this is that they lack control, are easily influenced by others. Healthy humans exert a higher control than children and handicapped people. You experience, or possess a greater extent of freedom proportional to the amount of influence and effectiveness of control you exert over the environment versus the environment over you. Though this definition of free will only works once you arbitrarily define a system to be an entity within an environment contrary to being the environment. Thus the neural activity, being either consciously aware and controlled by the system itself, or not, is no valid argument within this framework. Of course, in a strong philosophical sense this definition fails to address the nature of free will as we can do what we want but not chose what we want. But I think it might after all be a useful definition when it comes to science, psychology and law. It might also very well address our public understanding of being free agents.
I should have checked the lesswrong wiki before posting this. And of course read the mentioned posts here on lesswrong.com.
One of the easiest hard questions, as millennia-old philosophical dilemmas go. Though this impossible question is fully and completely dissolved on Less Wrong...
Anyway, for those who care or are wondering what I have been talking about I thought I should provide some background information. My above drivel is loosely based on work by Björn Brembs et al.
“Our results address the middle ground between simple determinism and randomness that is currently not well understood or characterized. We speculate that if free will exists, it is in this middle ground.” This leads me to believe that the question of whether or not we have free will appears to be posed the wrong way. Instead, if we ask ‘where between chance and necessity are we located?’ one finds that this is precisely where humans and animals differ. Humans may not have free will in the philosophical sense, but even flies have a number of behavioral options they need to decide between. Humans are less determined than flies and possess even more options. With this small reformulation, the topic of free will becomes the new biological research area of studying spontaneous behavior and can thus be discerned from the philosophical question.
Maybe a misinterpretation on my side. But now my above comments might make a bit more sense, or at least show where I’m coming from. I learnt about this via a chat about ‘free will’.
Hope you don’t mind I post this. Maybe somebody will find it useful or informative.
Faced with novel situations, humans and most animals spontaneously increase their behavioural variability...
Controlling external events: the input
Thus, competitive success and evolutionary fitness of all ambulatory organisms rely critically on intact behavioral variability as an adaptive brain function. But relative freedom from environmental contingencies is a necessary, but most often not a sufficient criterion for such accomplishments. Tightly connected to the ability to produce variable behavior is the ability to use the effects of these behaviors to control the environment. The incoming stream of sensory information is noisy and fluctuates for any number of reasons. Any covariance between the behavioral variations and those of sensory input indicates that the latter are con-sequences of the behavior and can thus be controlled be the animal. This function is so paramount, that we humans express our delight over control of our envi-ronment (including other people) already as children, by e.g., shrieking in excitement when Daddy jumps after a “boo” or proudly presenting Mom with “look what I can do!”.
In my opinion you have made a rather egregious error in your evaluation of the issue of free will. You seem to have a dearly held pre-conceived notion that for anything to be established as true it must be proved in a laboratory. From which Mount Sinai did you receive this proclamation? In fact it is an article of faith.
The perception of every human being who has ever lived and is alive today tells us clearly that our actions are based on free will decisions. I can change the way I feel, I can change the way I behave by exercising my free will. I can decide what I want to think about and when I want to think about it. I can decide whether to shut off the alarm and go back to sleep or get out of bed early and do my daily exercise regimen.
if “Science” doubts the existence of free will then there is something wrong with Science not my clear perception of my free will (along with the clear perceptions of just about everyone alive and who has ever lived). It is your problem to “prove” that free will does not exist, not my problem to prove that it does exist.
One thing I do agree with; Free will is something that is beyond the material world.
But of course we are involved in non-material (supernatural, spiritual, etc.) activities from the moment we wake up until the moment we go to sleep. Communicating in written language like everyone on this blog is doing is one of them. We type absolutely meaningless symbols on a screen and somehow the ideas in my head get conveyed to whoever “reads” them. Take the most advanced laboratory in the world, and have them analyze the ink on a piece of paper and the paper itself. The laboratory can tell you everything about the chemical and molecular structure of both, but it cannot hope to ever figure out the message that is written there, AND YET IT IS THERE NONETHELESS. We attach non-material ideas to meaningless symbols on a piece of paper or on a computer screen.
if “Science” doubts the existence of free will then there is something wrong with Science not my clear perception of my free will (along with the clear perceptions of just about everyone alive and who has ever lived).”
Whatever the conclusion here, it will not be attained by ninja’ing “everybody’s perceptions” into natural law. Everybody perceives that heavy objects fall faster than light ones. Everybody perceives that squares A and B are different colours.
Do I need to recite the whole litany? We aren’t Aristotelians or apologists or something, we don’t get to do philosophy just by sitting back in our armchairs and imagining how the world “must” obviously be. You have to actually go and look at the world. Hence our “faith” in the lab.
Perhaps I should state it in a slightly different way. There is no reason for me or anyone else to doubt the clear perception we have of our own free will. Prove to me scientifically that it does not exist, that it is some kind of illusion that all human beings experience.
Perhaps I should state it in a slightly different way. There is no reason for me or anyone else to doubt the clear perception we have of our own free will. Prove to me scientifically that it does not exist, that it is some kind of illusion that all human beings experience.
Fair enough. I’m glad you agree it’s an empirical question.
Having got that concession from you, I’ll tell you I actually don’t agree with Chu-Carroll’s analysis. Of course everything is definition-dependent here, but in essence, I don’t think free will is an illusion. Rather, I think the opposition of determinism to free will is just mistaken. Determinism does not imply no free will. This position is called compatibilism.
What Chu-Carroll is saying is that free will is not some weird force outside physics that thaumaturgically makes an electron zig instead of zag, causing the miracle of choice. I agree up to there. So if that’s what you call free will, then it is an illusion. He then implies “there is no free will.” Indeed, not under that definition.
But that’s not what I call free will.
See Daniel Dennett’s “Elbow Room;” also search this site for “Free Will.” Eliezer has done some excellent writing on the subject.
Something is either determined or it is undetermined, to some degree random. We can make sense of no third option. The free will you want is apparently not compatible with determinism (too bad, mine is). But the free will you want is also not random- how could we be held responsible for a random event? Could a flip of a coin be what determines whether you act wrongly or rightly?
You ask for something that is neither determined nor undetermined and such a thing is impossible on pain of sacrificing the basic concepts we use to understand the world.
The perception of every human being who has ever lived and is alive today tells us clearly that our actions are based on free will decisions. I can change the way I feel, I can change the way I behave by exercising my free will. I can decide what I want to think about and when I want to think about it. I can decide whether to shut off the alarm and go back to sleep or get out of bed early and do my daily exercise regimen.
Of course we can. But that isn’t evidence of the sort of free will you’re talking about. I can do whatever I want. It just so happens that what I want is causally determined. That’s okay, almost everything is causally determined. The kind of free will you’re talking about isn’t even magic. At least everything Harry Potter, Santa Clause and Jesus do is conceptually coherent. Jesus didn’t turn water into square circles!
You ask for something that is neither determined nor undetermined and such a thing is impossible on pain of sacrificing the basic concepts we use to understand the world.
Can you say a bit more about that? I dislike disturbing other’s basic concepts of world-understanding without good reason, but my basic categories are simply threefold: it seems basic to me that events can be caused, chosen, or random.
To hear that it is impossible for something to be non-caused and non-random at the same time feels to me a bit like hearing that it’s impossible for a color with a pure hue to be non-red and non-green; the answer is simply that the color must be blue.
it seems basic to me that events can be caused, chosen, or random.
Can you give a specific, meaningful definition of “choice” that cannot be reduced to causality, randomness, or some mixture thereof? I’ve tried, and I really can’t think of any that hold up under scrutiny. Most of my attempts (from back when this seemed like a serious problem to me) turned out to be meaningless or circular definitions, or homunculi. Even if, for example, you posit an extraphysical soul that controls one’s body like a puppet, then the question of “How does the soul make its choices?” is still meaningful — a given choice can either be for a reason or for no reason — and there would in theory be a right answer, even if there we somehow could never find out what it was (and we probably could anyway, by studying behavior and reverse-engineering general principles: the same thing science always does, even if it cannot directly study the underlying mechanism).
No, I can’t. I think that, in practice, the three categories are hopelessly interwoven. You should be able to “reduce” any event into any of the three categories.
Take the thing about the wallet. I can focus on the choice I made, and say it was my choice that mattered; I could have chosen differently, and but for my choice, the wallet would not have been returned. I can focus on the randomness, and say it was the randomness that mattered; I happened to win a small prize on a scratch-off lottery earlier that day, and but for that randomness, the wallet would not have been returned. I can focus on the casuation, and say that it was the prior conditions that mattered; a perfect computer could have known that I had brain-state X immediately before returning the wallet; and brain-state X is sufficient to induce wallet returns.
Well, why did I have brain-state X? It could be because I had brain-state Y a moment ago, which is a sufficient cause of X, or it could be because the electrons in some of my neurons randomly happened to be in the right place at the right time, or it could be because I chose to look down at the ground and see what was there.
And so on—any given event can be explained in any of three different ways; when it comes to anything as complex as the human brain, the pure types exist only in our imaginations.
I assume that there is some line you would draw, be it at individual neurons, or cellular structures, or molecules, or individual atoms, beyond which you would say, “these types of things don’t make choices.” If so, how would you reply to the question: in what sense can a system built out of components that don’t make choices be said to make choices?
[grin] Cyan, I actually do believe that even subatomic particles, in some limited sense, can be said to make choices. I do agree with you that choice can’t arise out of choiceless components.
If you’re curious about what it might mean for an electron to make a choice, and you have a high tolerance for whimsy, I recommend the book Reenchantment without Supernaturalism. The philosophy isn’t very rigorous, and it’s rather out of date, but it’s the only one I know of that (a) takes physics seriously, (b) takes logic seriously, and (c) accounts for my intuition that I make choices without (d) dismissing that intuition as an illusion.
Quantum amplitudes evolve deterministically, and it’s generally held that quantum systems either decohere deterministically or collapse randomly. How does this permit subatomic particles to be said to make choices, even in a limited sense?
Is there a more easily accessible explanation of the argument in the book? (I’m not going to shell out for it.)
Not that I know of, although it might be in a good public library. Sorry about that; I know it’s unfair to ask you to look in an obscure book to get the basic drift of a contrarian argument; that sort of thing is rarely worth people’s time. If you happen to live near Boston, Miami, or San Francisco, which are places I’ll be over the next few months, then I’ll be happy to lend you my copy.
I don’t know nearly enough quantum physics to give you an intelligent answer to your question about amplitudes and systems in my own words. Again, sorry about that.
If you happen to live near Boston, Miami, or San Francisco...
Nope. Oh well, at least the exchange served to clarify that you really do consider choice ontologically fundamental, and not just a useful category for practical purposes.
You might have to answer ata’s question before I can say more. But try this. Take a stock example of a free choice. Smith finds a wallet on the ground. He decides to return it. Why did he return it? Now maybe Smith didn’t choose to return it, maybe someone made him do it by hypnotizing him. This is obviously a causal event, presumably we can agree that Smith isn’t using libertarian free will here. But maybe he did choose to return it. Even then though we don’t just stop wonder “why?”. When we ask for an explanation of behavior “I chose to” isn’t enough. Usually we ask for reasons. Reasons are significant explanations for actions because they appear to be morally significant. We don’t usually hold drugged or (were it to happen) mind-controlled people responsible for their actions to the same degree people he choose to act based on reasons. But reasons are still causal explanations. Were returning the wallet not the moral thing to do (say the wallet had proof the owner had committed a heinous crime) then the Smith would have no returned the wallet. Explanations based on reasons can be reduced to sentences about neurons and biochemistry the same way any causal explanation can. The only explanation that I can think of that wouldn’t be a causal explanation is if the choice to return the wallet just happened. There was no reason, no cause. It just happened. It could have not happened but in did. Thats what we mean by undetermined or random. There is a middle ground, some things can be more likely than other things. Some philosophers have tried to argue that brains are sensitive enough to quantum level indeterminacy that our actions cannot be perfectly predicted even in principle. But even if this is why Smith returned the wallet it obviously doesn’t give Smith the kind of free will libertarians want.
I just don’t know what this other thing could be. Determined and indeterminate appear to exhaust the possibility space. Philosophers have tried for a very long time to offer something else up but they inevitably fail.
The only explanation that I can think of that wouldn’t be a causal explanation is if the choice to return the wallet just happened. There was no reason, no cause.
Just so. If I choose to return a wallet, that is just a brute fact about the universe. You should feel free to ask the question “Why?” about a pure choice, but I cannot think of any good answers to it. The only way you can get any analytical traction over choices is if you have more than one choice, or if the choices are mixed in with some randomness or some causality.
There was no reason, no cause. It just happened. It could have not happened but in did. That’s what we mean by undetermined or random.
Your description is necessary for X to be random, but not sufficient for X to be random. Your description, to me, is just another way of saying that X is not caused. If you think that all things are either caused or random, then you will naturally conclude that a non-caused thing is random. If, however, you start off thinking that all things are caused, random, or chosen, then you will naturally conclude that a non-caused thing is either random or chosen—it will not occur to you to assume that all non-caused things must be random.
Conversely, what I mean by random is slightly more specific than what you mean by random. When you say “random,” you just mean that something is not caused. When I say “random,” I mean that something is not caused AND not chosen.
Randomness is a brute fact about the universe. We are always probing to find out what really underlies a probability distribution, but whenever we pause in our labors and treat a distribution as if it were actually driven by randomness, there is nothing interesting that we can say about that randomness. Why did the coin come up heads instead of tails? I have no idea. I can wave my hand in the direction of atmospheric physics and conservation of angular momentum, but we both know that I don’t have anywhere near enough information or computing power to actually calculate the path of the coin through the air.
Similarly, choice is a brute fact about the universe. When you ask me why I returned the wallet, I can wave my hand in the direction of ethics and psychology, but we both know that I don’t have anywhere near enough self-understanding or computing power to actually calculate the roots of my decision in my past.
The only difference between the coin and the wallet is that I controlled the path of the wallet in a way I did not control the path of the coin. They are both “random” in the sense of being un-caused, but one is “nonrandom” in the sense of being chosen.
If I choose to return a wallet, that is just a brute fact about the universe.
So I made this particular argument partly because it’s one of my favorites, succinct and devastating but also because going into what we know about brain science with the troll above would be useless. But lets go into it for a minute. We’re far from understanding everything about the brain but we know a lot and have good reason to think that everything we do is the product of neuron firings. How the brain works (making choices and otherwise) can be complicated but it is almost definitely causal. As I said some have tried to show that they are random… but what is the third way brains could be? If choice is basic you’re saying it can’t be reduced to neuron firings, but everything we know about the brain suggests it can. Where is choice in the objective realm, the world of particles and forces?
Randomness is a brute fact about the universe.
Well in Quantum mechanics it might be but
Why did the coin come up heads instead of tails? I have no idea. I can wave my hand in the direction of atmospheric physics and conservation of angular momentum, but we both know that I don’t have anywhere near enough information or computing power to actually calculate the path of the coin through the air.
The fact that your or I are not capable of calculating the path of a flipping coin doesn’t make randomness basic. The fact that in principle we could calculate it given enough information and brain power proves that the path of a coin is actually causally determined. My mention of a coin flip in my original comment was metaphorical.
Similarly, choice is a brute fact about the universe. When you ask me why I returned the wallet, I can wave my hand in the direction of ethics and psychology, but we both know that I don’t have anywhere near enough self-understanding or computing power to actually calculate the roots of my decision in my past.
Your ignorance is not cause for postulating fundamental ontologies. It is a mysterious answer. If we knew more about ethics and psychology you absolutely could do this calculation. Besides which, in many circumstances you know exactly why someone made a choice. I don’t need to know anything about psychology or brains to know that someone who returns a wallet is probably doing so because they are being ethical or hoping for a reward. These are obvious reasons for choosing to return a wallet.
Also please distinguish choice from randomness. I’m still not quite sure what this concept consists of. How can something be uncaused but not random? If there is no cause what constrains the outcome? If there is no constraint on the outcome how is it not random?
We’re far from understanding everything about the brain but we know a lot and have good reason to think that everything we do is the product of neuron firings.
Neurologically speaking, I’ll admit that you could be right. It could turn out to be the case that humans happen to be a sort of creature that makes decisions based on neuron firings, and that neuron firings are in turn based entirely on deterministic particle collisions plus a bit of quantum randomness on the side. I keep half an eye on the neuroscience articles in the popular press, and if and when they report that conclusion, I’ll take it seriously. It would force me to revise several of my core beliefs. One belief that wouldn’t change, though, is the belief that, as a matter of philosophy, there’s nothing wrong with the idea of a choosing being, even if we have no real life examples of choosing beings on Earth.
Your ignorance is not cause for postulating fundamental ontologies. It is a mysterious answer. If we knew more about ethics and psychology you absolutely could do this calculation.
I’ve read and re-read the page on mysterious answers, and I think it’s a great article. I don’t mean to shut off inquiry at all by saying that I choose things; by all means, scan my brain and tell me what you see! I’ll be mildly curious. I’m not curious enough to do the scans myself, because I’m too busy trying to investigate my mind at the level of macrophenomena like “habits,” “willpower,” and “awareness” to bother much with the microfoundations of those phenomena in individual neurons or clusters. I expect that within my lifetime, brain scans will be cheap, safe, and precise enough that I’ll be able to get better information from physical science than from introspection, and then I’ll switch the bulk of my investigative activity over to physical science. In the meantime, I find that “choice” works just fine as a placeholder in the heuristic equations I use to model my mental macrophenomena. It may or may not correspond to anything real at the quantum level, but it helps me understand myself, so I’m using it.
Also please distinguish choice from randomness. I’m still not quite sure what this concept consists of. How can something be uncaused but not random? If there is no cause what constrains the outcome? If there is no constraint on the outcome how is it not random?
I’m sorry; I’m not sure what else to say. if my analogy about red/blue/green and my necessary vs. sufficient paragraph didn’t get the point across, then I don’t know how else to explain it. if you insist on using a mental model that has only two possible values for a variable, then talk about a third value will not make any sense to you; I cannot stop you from trying to explain the third value in terms of your existing mental model and then getting confused or annoyed when it doesn’t work.
Neurologically speaking, I’ll admit that you could be right. It could turn out to be the case that humans happen to be a sort of creature that makes decisions based on neuron firings, and that neuron firings are in turn based entirely on deterministic particle collisions plus a bit of quantum randomness on the side. I keep half an eye on the neuroscience articles in the popular press, and if and when they report that conclusion, I’ll take it seriously.
But this is what is so great about Bayesian epistemology. You don’t have to wait for some neuroscientists to announce this finding. If you know a decent amount of neuroscience now, you can be fairly confident in predicting that they one day will be able to explain choice in terms of neuron firings. All the people here who believe this aren’t just making it up. We’re extrapolating from what is known and making reasonable inferences. If you wait for someone to figure out exactly how it is done you’re going to spend a lot more time being wrong than those who infer in advance. Again though, I can already make accurate predictions about people’s choices based on macro-phenomena.
In the meantime, I find that “choice” works just fine as a placeholder in the heuristic equations I use to model my mental macrophenomena. It may or may not correspond to anything real at the quantum level, but it helps me understand myself, so I’m using it.
But don’t confuse placeholders with fundamental properties. I have no problem with “choice”. I use it all the time. I think I make choices constantly. If it is helpful for your models by all means use it. But that doesn’t require you to assert that choice is some incredible new kind of event which is neither causal nor random. I have lots of things in my ontology that are not in my basic ontology: morality, love, basketball etc. Maybe in modeling subjective experience you even want to distinguish things you do from other caused or random events and so use this word “choice” in a special way. But surely you can recognize that you aren’t actually that different from all the other objects you discover in the world and likely work the same way they do. And when you take this objective, view from nowhere, scientific perspective I don’t see how you can have an event that is neither caused nor random.
’m sorry; I’m not sure what else to say. if my analogy about red/blue/green and my necessary vs. sufficient paragraph didn’t get the point across, then I don’t know how else to explain it. if you insist on using a mental model that has only two possible values for a variable, then talk about a third value will not make any sense to you; I cannot stop you from trying to explain the third value in terms of your existing mental model and then getting confused or annoyed when it doesn’t work.
I quite understand that your “choice” is neither caused or random but a third value that is neither. What I don’t understand is what positive qualities this third value possesses. I promise you I can make sense of their being a third variable in abstract. What I don’t understand is what your third variable is. I can say lots of things about “random” and “causally determined” that distinguish these properties. But I haven’t heard you do anything in the way of describing this third property.
You should know that you’re hardly the first person who has wanted this kind of free will and went about inventing a third kind of thing to prove that it existed. One reason I’m so skeptical is that every single one of these attempts that I know of has failed miserably. Libertarians are a very small minority among contemporary analytic philosophers for a reason.
I quite understand that your “choice” is neither caused or random but a third value that is neither.
OK, good, I thought so. You seemed pretty smart.
I can say lots of things about “random” and “causally determined” that distinguish these properties.
Why don’t you go ahead and do that, for a paragraph or so, and I’ll see if I can complete the pattern for you and give you the kind of description you’re looking for. To me it just seems obvious what a choice is, in the same way that I know what “truth” is and what “good” is, but if you can manage to describe the meaning of “random” analytically then I can probably copy it for the word “chosen.” If I can’t, that will surprise me.
But surely you can recognize that you aren’t actually that different from all the other objects you discover in the world and likely work the same way they do.
you’re hardly the first person who has wanted this kind of free will and went about inventing a third kind of thing to prove that it existed.
Have I waxed poetic about souls and destiny and homunculi? I don’t remember “inventing” a third kind of thing. I’m just sort of pointing at my experience of choice and labeling it “choice.” If you insist that what I think is choice is really something else, you’re welcome to prove it to me with direct evidence, but I’m not really interested in Bayesian inferences here. I am unconvinced that brains and rocks are in the same reference class. I do not accept the physicalist-reductionist hypothesis as literally true, despite its excellent track record at producing useful models for predicting the future. I understand that the vast majority of people on this site -do- accept that hypothesis. I do not have the stamina or inclination to hold the field on that issue against an entire community of intelligent debaters.
Why don’t you go ahead and do that, for a paragraph or so, and I’ll see if I can complete the pattern for you and give you the kind of description you’re looking for. To me it just seems obvious what a choice is, in the same way that I know what “truth” is and what “good” is, but if you can manage to describe the meaning of “random” analytically then I can probably copy it for the word “chosen.” If I can’t, that will surprise me.
It’s obvious to you what “truth” is and what “goodness” is? Really? I think I can say clever and right things about these concepts because I’ve done a lot of studying and thinking. But the answers don’t seem obvious at all to me. Anyway, causality and randomness. Clearly huge topics about which lots have been said.
I believe a causal event is a kind of regularity, extended in spacetime, which has a variable that can be manipulated by hypothetical agent at one end to control a variable at the other end (usually the effect part is later in time). So by altering the velocity of an asteroid, the mean temperature of the planet Earth can be dramatically altered, for example. On a micro-level, intervening on a neuron and causing it to fire at a certain rate will lead to adjacent neurons firing. Altering the social mores of a society can cause a man not to return a wallet. For any one event to occur a large amount of variable have to be right and any one of those variable can be altered so as to alter the event, so these simple examples are overly simple. Lots more has been said if you’re interested. Pearl and Woodward are good authors.
Randomness might be more difficult since it isn’t obvious ontological randomness even exists. Epistemological randomness does: rolling a dice is a good example we have no way to predict the outcome. But in principle we could predict the outcome. Some interpretations of quantum mechanics do involve ontological randomness. Such events can be distinguished from causal events in that the valuable of the resulting variable cannot be controlled by any agent, not because no agent is powerful enough but because there are no variables which can be intervened on to alter the outcome in the way desired. There is no possibility of controlling such events. It is possible quantum indeterminacy is just the product of a hidden variable we don’t know about or that the apparent randomness is actually just a product of anthropics, every possible state gets observed and every outcome seems random because “you” only get to observe one and can’t communicate with the other “you’s”.
Have I waxed poetic about souls and destiny and homunculi? I don’t remember “inventing” a third kind of thing. I’m just sort of pointing at my experience of choice and labeling it “choice.” If you insist that what I think is choice is really something else, you’re welcome to prove it to me with direct evidence, but I’m not really interested in Bayesian inferences here.
I don’t have a problem with you pointing at an experience and labeling it “choice”. I do that too. You make choices. It’s just what it is to make a choice is one of these two things, a caused event or an uncaused event. You invent a third kind of thing when you come up with with a new kind of event which isn’t seen anywhere else, and declare it to be fundamental. And the way many philosophers have historically dealt with this exact problem is by positing souls and homunculi, “agent causation” and whatnot. When you decide that your experience of choice is a fundamental feature of the world you’re doing the exact same thing- any claim that something is irreducible is the same as a claim that something belongs in our basic ontology. The fact that you didn’t do this in verse just means I’m not annoyed, it’s still the same mistake.
I am unconvinced that brains and rocks are in the same reference class. I do not accept the physicalist-reductionist hypothesis as literally true, despite its excellent track record at producing useful models for predicting the future. I understand that the vast majority of people on this site -do- accept that hypothesis. I do not have the stamina or inclination to hold the field on that issue against an entire community of intelligent debaters.
I’ve been known to be more tolerant that others of unorthodoxy on this matter and I doubt many more would join in. Most people probably have the same arguments anyway. You’re not obligated to but I’d be interested in hearing your reasons for not accepting the hypothesis. However, my definition of truth is something like “the limit of useful modeling” so we might have to sort truth out a bit too. If you preface the discussion to demonstrate that you’re aware the position is unpopular already and you’re just trying to work this out you can probably avoid a karma hit. I’ll vote you up it it happens.
f you preface the discussion to demonstrate that you’re aware the position is unpopular already and you’re just trying to work this out you can probably avoid a karma hit.
Sure, consider it prefaced. I’m not trying to convince anybody; I’m just sharing my views because one or two users seem curious about them, and because I might learn something this way. It’s not very important to me. If anyone would like me to stop talking about this topic on Less Wrong, feel free to say so explicitly, and I will be glad to oblige you.
It’s obvious to you what “truth” is and what “goodness” is? Really?
I don’t mean that the entire contents, in detail, of what is and is not inside the box marked “true” is known to me. That would be ridiculous. I just mean that I know which box I’m talking about, and so do you. Sophisticated discussions about what “true” means (as opposed to discussion about whether some specific claim X is true) generally do more harm than good. You can tell cute stories about The Simple Truth, and that may help startle some philosophers into realizing where they’ve gone off-course, but mostly you’re just lending a little color to the Reflexive Property or the Identity Property: a = a.
Some interpretations of quantum mechanics do involve ontological randomness. Such events can be distinguished from causal events in that the valuable of the resulting variable cannot be controlled by any agent, not because no agent is powerful enough but because there are no variables which can be intervened on to alter the outcome in the way desired. There is no possibility of controlling such events.
I can probably work with this. I expect you will still think I’m postulating unnecessary ontological entities, and, given your epistemological value system, you’ll be right. Still, maybe the details will interest you.
Some interpretations of conscious awareness do involve ontological choice. Such events can be distinguished from random events in that the value of the resulting variable can be controlled by exactly one agent, as opposed to zero agents, as in the case of a truly random variable. The agent in question could be taken to be some subset of the neurons in the brain, or some subset of a person’s conscious awareness, or some kind of minimally intervening deity. It is not clear exactly who or what the agent is.
Conscious events can be distinguished from caused events in that conventional measures of kinetic power and information-theoretic power are bad predictors of a hypothetical agent’s ability to manipulate the outcome of a conscious event. Whether because the relevant interactions among neurons, given their level of chaotic complexity, occur in a slice of spacetime that is small enough to be resistant to external computation, or because the event is driven by some process outside the well-understood laws of physics, a conscious event is difficult or impossible to control from outside the relevant consciousness. Thus, instead of a single output depending subtly on many other variables, the output depends almost exclusively on a single input or small set of inputs.
You’re not obligated to but I’d be interested in hearing your reasons for not accepting the hypothesis.
I’d be happy to explain it in August, when I’ll be bored silly. At the moment, I’m pretty busy with my law school thesis, which is on antitrust law and has little to do with either free will or reductionism. Feel free to comment on any of my posts around that time, or to send your contact info to zelinsky a t gm ail dot com. Zelinsky is a rationalist friend of mine who agrees with you and only knows one person who thinks like me, so he’ll know who it’s for.
Thanks for bearing with me so far and for responding to arguments that must no doubt strike you as woefully unenlightened with a healthy measure of respect and patience. I really am done with both the free will discussion and the reductionist discussion for now, but I enjoyed discussing them with you, and consider it well worth the karma I ‘spent’. If you can think of any ways that what you see as my misunderstanding of free will or reductionism is likely to interfere with my attempts to help refine LW’s understanding of Goodhart’s Law, please let me know, and I’ll vote them up.
For convenience. If you show me a few examples where believing that I don’t have free will helps me get what I want, I might start caring about the actual structure of my mental algorithms as seen from the outside.
It is beneficial to believe you don’t have free will if you don’t have free will. From Surely You’re Joking, Mr. Feynman!:
When the real demonstration came he had us walk on stage, and he hypnotized us in front of the whole Princeton Graduate College. This time the effect was stronger; I guess I had learned how to become hypnotized. The hypnotist made various demonstrations, having me do things that I couldn’t normally do, and at the end he said that after I came out of hypnosis, instead of returning to my seat directly, which was the natural way to go, I would walk all the way around the room and go to my seat from the back.
All through the demonstration I was vaguely aware of what was going on, and cooperating with the things the hypnotist said, but this time I decided, “Damn it, enough is enough! I’m gonna go straight to my seat.”
When it was time to get up and go off the stage, I started to walk straight to my seat. But then an annoying feeling came over me: I felt so uncomfortable that I couldn’t continue. I walked all the way around the hall.
I was hypnotized in another situation some time later by a woman. While I was hypnotized she said, “I’m going to light a match, blow it out, and immediately touch the back of your hand with it. You will feel no pain.”
I thought, “Baloney!” She took a match, lit it, blew it out, and touched it to the back of my hand. It felt slightly warm. My eyes were closed throughout all of this, but I was thinking, “That’s easy. She lit one match, but touched a different match to my hand. There’s nothin’ to that; it’s a fake!”
When I came out of the hypnosis and looked at the back of my hand, I got the biggest surprise: There was a burn on the back of my hand. Soon a blister grew, and it never hurt at all, even when it broke.
So I found hypnosis to be a very interesting experience. All the time you’re saying to yourself, “I could do that, but I won’t”—which is just
another way of saying that you can’t.
All right, suppose all that is true, and that people can be hypnotized so that they literally can’t break away from the hypnotizing effect until released by the hypnotist.
That suggests that I should believe that hypnotism is dangerous. It would be useful to be aware of this danger so that I can avoid being manipulated by a malicious hypnotist, since it turns out that what appears to be parlor tricks are actually mind control. Great.
But, if I understand it correctly, which I’m not sure that I do, a world without free will is like a world where we are always hypnotized.
Once you’re under the hypnotist’s spell, it doesn’t do any good to realize that you have no free will. You’re still stuck. You will still get burned or embarrassed if the hypnotist wants to burn you.
So if I’m already under the “hypnotist’s” spell, in a Universe where the hypnotist is just an impersonal combination of an alien evolution process and preset physical constants, why would I want to know that? What good would the information do me?
I’m sorry, I’m not maintaining that free will is incompatible with determinism, only that sometimes free will is not present, even though it appears to be. When hypnotized, Richard Feynman did not have (or, possibly, had to a greatly reduced extent) free will in the sense that he had free will under normal circumstances—and yet subjectively he noticed no difference.
It appears to me that you created your bottom line from observing your subjective impression of free will. I suggest that you strike out the entire edifice you built on these data—it is built on sand, not stone.
I see; I did misunderstand, but I think I get your point now. You’re not claiming that if only Mr. Feynman had known about the limits of free will he could have avoided a burn; you’re saying that, like all good rationalists everywhere, I should only want to believe true things, and it is unlikely that “I have free will” is a true thing, because sometimes smart people think that and turn out to be wrong.
Well, OK, fair enough, but it turns out that I get a lot of utility out of believing that I have free will. I’m happy to set aside that belief if there’s some specific reason why the belief is likely to harm me or stop me from getting what I want. One of the things I want is to never believe a logically inconsistent set of facts, and one of the things I want is to never ignore the appropriately validated direct evidence of my senses. That’s still not enough, though, to get me to “don’t believe things that have a low Bayesian prior and little or no supporting evidence.” I don’t get any utility out of being a Bayesianist per se; worshipping Bayes is just a means to an end for me, and I can’t find the end when it comes to rejecting the hypothesis of free will.
Robin, I’ve liked your comments both on this thread and others that we’ve had, but I can’t afford to continue the discussion any time soon—I need to get back to my thesis, which is due in a couple of weeks. Feel free to get in the last word; I’ll read it and think about it, but I won’t respond.
My last word, as you have been so generous as to give it to me, is that I actually do think you have free will. I believe you are wrong about what it is made of, just as the pre-classical Greeks were wrong about the shape of the Earth, but I don’t disagree that you have it.
Good luck on your thesis—I won’t distract you any more.
I place a very low probability on my having genuine ‘free will’ but I act as if I do because if I don’t it doesn’t matter what I do. It also seems to me that people who accept nihilism have life outcomes that I do not desire to share and so the expected utility of acting as if I have free will is high even absent my previous argument. It’s a bit of a Pascal’s Wager.
Why do you define “free will” to refer to something that does not exist, when the thing which does exist—will unconstrained by circumstance or compulsion—is useful to refer to? For one, its absence is one indicator of an invalid contract.
I’m not exactly sure what you’re accusing me of. I think Freedom Evolves is about the best exposition of how I conceive of free will. I am also a libertarian. I find it personally useful to believe in free will irrespective of arguments about determinism and I think we should have political systems that assume free will. I still have some mental gymnastics to perform to reconcile a deterministic material universe with my own personal intuitive conception of free will but I don’t think that really matters.
I don’t really understand what you mean when you use the word ‘libertarian’ - it doesn’t seem particularly related to my understanding. I mean it in the political sense. Perhaps there is a philosophical sense that you are using?
Libertarian is the name for someone who believes free will exists and that free will is incompatible with determinism. Lol, it didn’t even occur to me you could be talking about politics.
Ok, I’ve done some googling and think I understand what you meant when you used the word. I’d never heard it in that context before. I guess philosophically I’m something like a compatibilist then, but I’m more of an ’it’s largely irrelevant’ist.
But, if I understand it correctly, which I’m not sure that I do, a world without free will is like a world where we are always hypnotized.
No! A world without libertarian free will is a world exactly like this one.
ETA: Robin’s point, I gather, is that a world without libertarian free will is a world where hypnotism is possible. Which, as it turns out, is this world.
I was actually making a lesser point: that the introspective appearance of free will is not even a reliable indicator of the presence of free will, much less a reliable guide to the nature of free will.
Edit: From which your interpretation follows, I suppose.
Can you say a bit more about that? I dislike disturbing other’s basic concepts of world-understanding without good reason, but my basic categories are simply threefold: it seems basic to me that events can be caused, chosen, or random.
(An aside: this sort of view is common in the free will dialectic—it is an incompatibilist theory, probably of the non-causal type. Jack’s objection is a standard one, and for good reason, I believe.)
Supernatural free will doesn’t exist. If your concept of free will requires it to be supernatural (mine doesn’t) and you are sure you experience free will clearly there is something wrong with either your concept or your perception.
First of all, you cannot think when and about what you want. Same as you cannot want what you want. This leads to an infinite recursion. No system can understand itself for that the very understanding would evade itself forever. A bin trying to contain itself.
Anyway, there is no reason to go beyond physical, factual inquiry right now. You can assess your data with practicability. If a drug makes you think that you can fly you can jump from the next bridge and be brought back down to earth by reality.
Reality is not subject to interpretation, only description and abstraction. The fundamental nature of reality, its characteristics and qualities are absolute. Red is always red even when you call it green.
You don’t have to have a comprehensive grasp to determine the simplest of all conclusions: Something exists or it doesn’t. Either something is tangible or it doesn’t exist. My definition of subsistence is for something to have an influence on me. This also implies that something that exists can be subject to scientific inquiry. Something that exists can be assessed with practicability. It makes a difference.
If I cut my throat I may discover that I was dreaming or that I have been playing some advanced virtual reality game all along. Everything is possible. But right now there are safer and more promising options of gaining knowledge. How can I be sure? I can’t, but there is evidence which proved to be reliable so far. I have to suspect that it will continue to be reliable based on experiment and observation. That doesn’t make it the ultimate way of knowing or even a superior way but the best I know of at this time. And until I hit some hard barrier I do not have any good reason to try something else.
-- Mark Chu-Carroll
There is more here: Brains as output/input devices
Although I should note that I believe there to be phenomena that qualify to be defined as ‘free will’. Specifically endogenous processes generating behavioral variability and thus non-linearity. Especially if you can show that the complexity of transformation by which a system shapes the outside environment, in which it is embedded, does trump the specific effectiveness of the environmental influence on the defined system. In other words, mind over matter. You are able to shape reality more effectively and goal-oriented and thus, in a way, overcome its crude influence it exerts on you. For example, children and some mentally handicapped people are not responsible in same the way as healthy adults. They can not give consent or enter into legally binding contracts. One of the reasons for this is that they lack control, are easily influenced by others. Healthy humans exert a higher control than children and handicapped people. You experience, or possess a greater extent of freedom proportional to the amount of influence and effectiveness of control you exert over the environment versus the environment over you. Though this definition of free will only works once you arbitrarily define a system to be an entity within an environment contrary to being the environment. Thus the neural activity, being either consciously aware and controlled by the system itself, or not, is no valid argument within this framework. Of course, in a strong philosophical sense this definition fails to address the nature of free will as we can do what we want but not chose what we want. But I think it might after all be a useful definition when it comes to science, psychology and law. It might also very well address our public understanding of being free agents.
I should have checked the lesswrong wiki before posting this. And of course read the mentioned posts here on lesswrong.com.
Anyway, for those who care or are wondering what I have been talking about I thought I should provide some background information. My above drivel is loosely based on work by Björn Brembs et al.
PLoS ONE: Order in Spontaneous Behavior
Maybe a misinterpretation on my side. But now my above comments might make a bit more sense, or at least show where I’m coming from. I learnt about this via a chat about ‘free will’.
Hope you don’t mind I post this. Maybe somebody will find it useful or informative.
There is more here: Brains as output/input devices
In my opinion you have made a rather egregious error in your evaluation of the issue of free will. You seem to have a dearly held pre-conceived notion that for anything to be established as true it must be proved in a laboratory. From which Mount Sinai did you receive this proclamation? In fact it is an article of faith.
The perception of every human being who has ever lived and is alive today tells us clearly that our actions are based on free will decisions. I can change the way I feel, I can change the way I behave by exercising my free will. I can decide what I want to think about and when I want to think about it. I can decide whether to shut off the alarm and go back to sleep or get out of bed early and do my daily exercise regimen.
if “Science” doubts the existence of free will then there is something wrong with Science not my clear perception of my free will (along with the clear perceptions of just about everyone alive and who has ever lived). It is your problem to “prove” that free will does not exist, not my problem to prove that it does exist.
One thing I do agree with; Free will is something that is beyond the material world. But of course we are involved in non-material (supernatural, spiritual, etc.) activities from the moment we wake up until the moment we go to sleep. Communicating in written language like everyone on this blog is doing is one of them. We type absolutely meaningless symbols on a screen and somehow the ideas in my head get conveyed to whoever “reads” them. Take the most advanced laboratory in the world, and have them analyze the ink on a piece of paper and the paper itself. The laboratory can tell you everything about the chemical and molecular structure of both, but it cannot hope to ever figure out the message that is written there, AND YET IT IS THERE NONETHELESS. We attach non-material ideas to meaningless symbols on a piece of paper or on a computer screen.
Whatever the conclusion here, it will not be attained by ninja’ing “everybody’s perceptions” into natural law. Everybody perceives that heavy objects fall faster than light ones. Everybody perceives that squares A and B are different colours.
Do I need to recite the whole litany? We aren’t Aristotelians or apologists or something, we don’t get to do philosophy just by sitting back in our armchairs and imagining how the world “must” obviously be. You have to actually go and look at the world. Hence our “faith” in the lab.
Perhaps I should state it in a slightly different way. There is no reason for me or anyone else to doubt the clear perception we have of our own free will. Prove to me scientifically that it does not exist, that it is some kind of illusion that all human beings experience.
Fair enough. I’m glad you agree it’s an empirical question.
Having got that concession from you, I’ll tell you I actually don’t agree with Chu-Carroll’s analysis. Of course everything is definition-dependent here, but in essence, I don’t think free will is an illusion. Rather, I think the opposition of determinism to free will is just mistaken. Determinism does not imply no free will. This position is called compatibilism.
What Chu-Carroll is saying is that free will is not some weird force outside physics that thaumaturgically makes an electron zig instead of zag, causing the miracle of choice. I agree up to there. So if that’s what you call free will, then it is an illusion. He then implies “there is no free will.” Indeed, not under that definition.
But that’s not what I call free will.
See Daniel Dennett’s “Elbow Room;” also search this site for “Free Will.” Eliezer has done some excellent writing on the subject.
Something is either determined or it is undetermined, to some degree random. We can make sense of no third option. The free will you want is apparently not compatible with determinism (too bad, mine is). But the free will you want is also not random- how could we be held responsible for a random event? Could a flip of a coin be what determines whether you act wrongly or rightly?
You ask for something that is neither determined nor undetermined and such a thing is impossible on pain of sacrificing the basic concepts we use to understand the world.
Of course we can. But that isn’t evidence of the sort of free will you’re talking about. I can do whatever I want. It just so happens that what I want is causally determined. That’s okay, almost everything is causally determined. The kind of free will you’re talking about isn’t even magic. At least everything Harry Potter, Santa Clause and Jesus do is conceptually coherent. Jesus didn’t turn water into square circles!
Can you say a bit more about that? I dislike disturbing other’s basic concepts of world-understanding without good reason, but my basic categories are simply threefold: it seems basic to me that events can be caused, chosen, or random.
To hear that it is impossible for something to be non-caused and non-random at the same time feels to me a bit like hearing that it’s impossible for a color with a pure hue to be non-red and non-green; the answer is simply that the color must be blue.
Can you give a specific, meaningful definition of “choice” that cannot be reduced to causality, randomness, or some mixture thereof? I’ve tried, and I really can’t think of any that hold up under scrutiny. Most of my attempts (from back when this seemed like a serious problem to me) turned out to be meaningless or circular definitions, or homunculi. Even if, for example, you posit an extraphysical soul that controls one’s body like a puppet, then the question of “How does the soul make its choices?” is still meaningful — a given choice can either be for a reason or for no reason — and there would in theory be a right answer, even if there we somehow could never find out what it was (and we probably could anyway, by studying behavior and reverse-engineering general principles: the same thing science always does, even if it cannot directly study the underlying mechanism).
No, I can’t. I think that, in practice, the three categories are hopelessly interwoven. You should be able to “reduce” any event into any of the three categories.
Take the thing about the wallet. I can focus on the choice I made, and say it was my choice that mattered; I could have chosen differently, and but for my choice, the wallet would not have been returned. I can focus on the randomness, and say it was the randomness that mattered; I happened to win a small prize on a scratch-off lottery earlier that day, and but for that randomness, the wallet would not have been returned. I can focus on the casuation, and say that it was the prior conditions that mattered; a perfect computer could have known that I had brain-state X immediately before returning the wallet; and brain-state X is sufficient to induce wallet returns.
Well, why did I have brain-state X? It could be because I had brain-state Y a moment ago, which is a sufficient cause of X, or it could be because the electrons in some of my neurons randomly happened to be in the right place at the right time, or it could be because I chose to look down at the ground and see what was there.
And so on—any given event can be explained in any of three different ways; when it comes to anything as complex as the human brain, the pure types exist only in our imaginations.
I assume that there is some line you would draw, be it at individual neurons, or cellular structures, or molecules, or individual atoms, beyond which you would say, “these types of things don’t make choices.” If so, how would you reply to the question: in what sense can a system built out of components that don’t make choices be said to make choices?
[grin] Cyan, I actually do believe that even subatomic particles, in some limited sense, can be said to make choices. I do agree with you that choice can’t arise out of choiceless components.
If you’re curious about what it might mean for an electron to make a choice, and you have a high tolerance for whimsy, I recommend the book Reenchantment without Supernaturalism. The philosophy isn’t very rigorous, and it’s rather out of date, but it’s the only one I know of that (a) takes physics seriously, (b) takes logic seriously, and (c) accounts for my intuition that I make choices without (d) dismissing that intuition as an illusion.
Quantum amplitudes evolve deterministically, and it’s generally held that quantum systems either decohere deterministically or collapse randomly. How does this permit subatomic particles to be said to make choices, even in a limited sense?
Is there a more easily accessible explanation of the argument in the book? (I’m not going to shell out for it.)
Not that I know of, although it might be in a good public library. Sorry about that; I know it’s unfair to ask you to look in an obscure book to get the basic drift of a contrarian argument; that sort of thing is rarely worth people’s time. If you happen to live near Boston, Miami, or San Francisco, which are places I’ll be over the next few months, then I’ll be happy to lend you my copy.
I don’t know nearly enough quantum physics to give you an intelligent answer to your question about amplitudes and systems in my own words. Again, sorry about that.
Nope. Oh well, at least the exchange served to clarify that you really do consider choice ontologically fundamental, and not just a useful category for practical purposes.
You might have to answer ata’s question before I can say more. But try this. Take a stock example of a free choice. Smith finds a wallet on the ground. He decides to return it. Why did he return it? Now maybe Smith didn’t choose to return it, maybe someone made him do it by hypnotizing him. This is obviously a causal event, presumably we can agree that Smith isn’t using libertarian free will here. But maybe he did choose to return it. Even then though we don’t just stop wonder “why?”. When we ask for an explanation of behavior “I chose to” isn’t enough. Usually we ask for reasons. Reasons are significant explanations for actions because they appear to be morally significant. We don’t usually hold drugged or (were it to happen) mind-controlled people responsible for their actions to the same degree people he choose to act based on reasons. But reasons are still causal explanations. Were returning the wallet not the moral thing to do (say the wallet had proof the owner had committed a heinous crime) then the Smith would have no returned the wallet. Explanations based on reasons can be reduced to sentences about neurons and biochemistry the same way any causal explanation can. The only explanation that I can think of that wouldn’t be a causal explanation is if the choice to return the wallet just happened. There was no reason, no cause. It just happened. It could have not happened but in did. Thats what we mean by undetermined or random. There is a middle ground, some things can be more likely than other things. Some philosophers have tried to argue that brains are sensitive enough to quantum level indeterminacy that our actions cannot be perfectly predicted even in principle. But even if this is why Smith returned the wallet it obviously doesn’t give Smith the kind of free will libertarians want.
I just don’t know what this other thing could be. Determined and indeterminate appear to exhaust the possibility space. Philosophers have tried for a very long time to offer something else up but they inevitably fail.
Just so. If I choose to return a wallet, that is just a brute fact about the universe. You should feel free to ask the question “Why?” about a pure choice, but I cannot think of any good answers to it. The only way you can get any analytical traction over choices is if you have more than one choice, or if the choices are mixed in with some randomness or some causality.
Your description is necessary for X to be random, but not sufficient for X to be random. Your description, to me, is just another way of saying that X is not caused. If you think that all things are either caused or random, then you will naturally conclude that a non-caused thing is random. If, however, you start off thinking that all things are caused, random, or chosen, then you will naturally conclude that a non-caused thing is either random or chosen—it will not occur to you to assume that all non-caused things must be random.
Conversely, what I mean by random is slightly more specific than what you mean by random. When you say “random,” you just mean that something is not caused. When I say “random,” I mean that something is not caused AND not chosen.
Randomness is a brute fact about the universe. We are always probing to find out what really underlies a probability distribution, but whenever we pause in our labors and treat a distribution as if it were actually driven by randomness, there is nothing interesting that we can say about that randomness. Why did the coin come up heads instead of tails? I have no idea. I can wave my hand in the direction of atmospheric physics and conservation of angular momentum, but we both know that I don’t have anywhere near enough information or computing power to actually calculate the path of the coin through the air.
Similarly, choice is a brute fact about the universe. When you ask me why I returned the wallet, I can wave my hand in the direction of ethics and psychology, but we both know that I don’t have anywhere near enough self-understanding or computing power to actually calculate the roots of my decision in my past.
The only difference between the coin and the wallet is that I controlled the path of the wallet in a way I did not control the path of the coin. They are both “random” in the sense of being un-caused, but one is “nonrandom” in the sense of being chosen.
So I made this particular argument partly because it’s one of my favorites, succinct and devastating but also because going into what we know about brain science with the troll above would be useless. But lets go into it for a minute. We’re far from understanding everything about the brain but we know a lot and have good reason to think that everything we do is the product of neuron firings. How the brain works (making choices and otherwise) can be complicated but it is almost definitely causal. As I said some have tried to show that they are random… but what is the third way brains could be? If choice is basic you’re saying it can’t be reduced to neuron firings, but everything we know about the brain suggests it can. Where is choice in the objective realm, the world of particles and forces?
Well in Quantum mechanics it might be but
The fact that your or I are not capable of calculating the path of a flipping coin doesn’t make randomness basic. The fact that in principle we could calculate it given enough information and brain power proves that the path of a coin is actually causally determined. My mention of a coin flip in my original comment was metaphorical.
Your ignorance is not cause for postulating fundamental ontologies. It is a mysterious answer. If we knew more about ethics and psychology you absolutely could do this calculation. Besides which, in many circumstances you know exactly why someone made a choice. I don’t need to know anything about psychology or brains to know that someone who returns a wallet is probably doing so because they are being ethical or hoping for a reward. These are obvious reasons for choosing to return a wallet.
Also please distinguish choice from randomness. I’m still not quite sure what this concept consists of. How can something be uncaused but not random? If there is no cause what constrains the outcome? If there is no constraint on the outcome how is it not random?
Neurologically speaking, I’ll admit that you could be right. It could turn out to be the case that humans happen to be a sort of creature that makes decisions based on neuron firings, and that neuron firings are in turn based entirely on deterministic particle collisions plus a bit of quantum randomness on the side. I keep half an eye on the neuroscience articles in the popular press, and if and when they report that conclusion, I’ll take it seriously. It would force me to revise several of my core beliefs. One belief that wouldn’t change, though, is the belief that, as a matter of philosophy, there’s nothing wrong with the idea of a choosing being, even if we have no real life examples of choosing beings on Earth.
I’ve read and re-read the page on mysterious answers, and I think it’s a great article. I don’t mean to shut off inquiry at all by saying that I choose things; by all means, scan my brain and tell me what you see! I’ll be mildly curious. I’m not curious enough to do the scans myself, because I’m too busy trying to investigate my mind at the level of macrophenomena like “habits,” “willpower,” and “awareness” to bother much with the microfoundations of those phenomena in individual neurons or clusters. I expect that within my lifetime, brain scans will be cheap, safe, and precise enough that I’ll be able to get better information from physical science than from introspection, and then I’ll switch the bulk of my investigative activity over to physical science. In the meantime, I find that “choice” works just fine as a placeholder in the heuristic equations I use to model my mental macrophenomena. It may or may not correspond to anything real at the quantum level, but it helps me understand myself, so I’m using it.
I’m sorry; I’m not sure what else to say. if my analogy about red/blue/green and my necessary vs. sufficient paragraph didn’t get the point across, then I don’t know how else to explain it. if you insist on using a mental model that has only two possible values for a variable, then talk about a third value will not make any sense to you; I cannot stop you from trying to explain the third value in terms of your existing mental model and then getting confused or annoyed when it doesn’t work.
But this is what is so great about Bayesian epistemology. You don’t have to wait for some neuroscientists to announce this finding. If you know a decent amount of neuroscience now, you can be fairly confident in predicting that they one day will be able to explain choice in terms of neuron firings. All the people here who believe this aren’t just making it up. We’re extrapolating from what is known and making reasonable inferences. If you wait for someone to figure out exactly how it is done you’re going to spend a lot more time being wrong than those who infer in advance. Again though, I can already make accurate predictions about people’s choices based on macro-phenomena.
But don’t confuse placeholders with fundamental properties. I have no problem with “choice”. I use it all the time. I think I make choices constantly. If it is helpful for your models by all means use it. But that doesn’t require you to assert that choice is some incredible new kind of event which is neither causal nor random. I have lots of things in my ontology that are not in my basic ontology: morality, love, basketball etc. Maybe in modeling subjective experience you even want to distinguish things you do from other caused or random events and so use this word “choice” in a special way. But surely you can recognize that you aren’t actually that different from all the other objects you discover in the world and likely work the same way they do. And when you take this objective, view from nowhere, scientific perspective I don’t see how you can have an event that is neither caused nor random.
I quite understand that your “choice” is neither caused or random but a third value that is neither. What I don’t understand is what positive qualities this third value possesses. I promise you I can make sense of their being a third variable in abstract. What I don’t understand is what your third variable is. I can say lots of things about “random” and “causally determined” that distinguish these properties. But I haven’t heard you do anything in the way of describing this third property.
You should know that you’re hardly the first person who has wanted this kind of free will and went about inventing a third kind of thing to prove that it existed. One reason I’m so skeptical is that every single one of these attempts that I know of has failed miserably. Libertarians are a very small minority among contemporary analytic philosophers for a reason.
OK, good, I thought so. You seemed pretty smart.
Why don’t you go ahead and do that, for a paragraph or so, and I’ll see if I can complete the pattern for you and give you the kind of description you’re looking for. To me it just seems obvious what a choice is, in the same way that I know what “truth” is and what “good” is, but if you can manage to describe the meaning of “random” analytically then I can probably copy it for the word “chosen.” If I can’t, that will surprise me.
Have I waxed poetic about souls and destiny and homunculi? I don’t remember “inventing” a third kind of thing. I’m just sort of pointing at my experience of choice and labeling it “choice.” If you insist that what I think is choice is really something else, you’re welcome to prove it to me with direct evidence, but I’m not really interested in Bayesian inferences here. I am unconvinced that brains and rocks are in the same reference class. I do not accept the physicalist-reductionist hypothesis as literally true, despite its excellent track record at producing useful models for predicting the future. I understand that the vast majority of people on this site -do- accept that hypothesis. I do not have the stamina or inclination to hold the field on that issue against an entire community of intelligent debaters.
It’s obvious to you what “truth” is and what “goodness” is? Really? I think I can say clever and right things about these concepts because I’ve done a lot of studying and thinking. But the answers don’t seem obvious at all to me. Anyway, causality and randomness. Clearly huge topics about which lots have been said.
I believe a causal event is a kind of regularity, extended in spacetime, which has a variable that can be manipulated by hypothetical agent at one end to control a variable at the other end (usually the effect part is later in time). So by altering the velocity of an asteroid, the mean temperature of the planet Earth can be dramatically altered, for example. On a micro-level, intervening on a neuron and causing it to fire at a certain rate will lead to adjacent neurons firing. Altering the social mores of a society can cause a man not to return a wallet. For any one event to occur a large amount of variable have to be right and any one of those variable can be altered so as to alter the event, so these simple examples are overly simple. Lots more has been said if you’re interested. Pearl and Woodward are good authors.
Randomness might be more difficult since it isn’t obvious ontological randomness even exists. Epistemological randomness does: rolling a dice is a good example we have no way to predict the outcome. But in principle we could predict the outcome. Some interpretations of quantum mechanics do involve ontological randomness. Such events can be distinguished from causal events in that the valuable of the resulting variable cannot be controlled by any agent, not because no agent is powerful enough but because there are no variables which can be intervened on to alter the outcome in the way desired. There is no possibility of controlling such events. It is possible quantum indeterminacy is just the product of a hidden variable we don’t know about or that the apparent randomness is actually just a product of anthropics, every possible state gets observed and every outcome seems random because “you” only get to observe one and can’t communicate with the other “you’s”.
I don’t have a problem with you pointing at an experience and labeling it “choice”. I do that too. You make choices. It’s just what it is to make a choice is one of these two things, a caused event or an uncaused event. You invent a third kind of thing when you come up with with a new kind of event which isn’t seen anywhere else, and declare it to be fundamental. And the way many philosophers have historically dealt with this exact problem is by positing souls and homunculi, “agent causation” and whatnot. When you decide that your experience of choice is a fundamental feature of the world you’re doing the exact same thing- any claim that something is irreducible is the same as a claim that something belongs in our basic ontology. The fact that you didn’t do this in verse just means I’m not annoyed, it’s still the same mistake.
I’ve been known to be more tolerant that others of unorthodoxy on this matter and I doubt many more would join in. Most people probably have the same arguments anyway. You’re not obligated to but I’d be interested in hearing your reasons for not accepting the hypothesis. However, my definition of truth is something like “the limit of useful modeling” so we might have to sort truth out a bit too. If you preface the discussion to demonstrate that you’re aware the position is unpopular already and you’re just trying to work this out you can probably avoid a karma hit. I’ll vote you up it it happens.
Sure, consider it prefaced. I’m not trying to convince anybody; I’m just sharing my views because one or two users seem curious about them, and because I might learn something this way. It’s not very important to me. If anyone would like me to stop talking about this topic on Less Wrong, feel free to say so explicitly, and I will be glad to oblige you.
I don’t mean that the entire contents, in detail, of what is and is not inside the box marked “true” is known to me. That would be ridiculous. I just mean that I know which box I’m talking about, and so do you. Sophisticated discussions about what “true” means (as opposed to discussion about whether some specific claim X is true) generally do more harm than good. You can tell cute stories about The Simple Truth, and that may help startle some philosophers into realizing where they’ve gone off-course, but mostly you’re just lending a little color to the Reflexive Property or the Identity Property: a = a.
I can probably work with this. I expect you will still think I’m postulating unnecessary ontological entities, and, given your epistemological value system, you’ll be right. Still, maybe the details will interest you.
Some interpretations of conscious awareness do involve ontological choice. Such events can be distinguished from random events in that the value of the resulting variable can be controlled by exactly one agent, as opposed to zero agents, as in the case of a truly random variable. The agent in question could be taken to be some subset of the neurons in the brain, or some subset of a person’s conscious awareness, or some kind of minimally intervening deity. It is not clear exactly who or what the agent is.
Conscious events can be distinguished from caused events in that conventional measures of kinetic power and information-theoretic power are bad predictors of a hypothetical agent’s ability to manipulate the outcome of a conscious event. Whether because the relevant interactions among neurons, given their level of chaotic complexity, occur in a slice of spacetime that is small enough to be resistant to external computation, or because the event is driven by some process outside the well-understood laws of physics, a conscious event is difficult or impossible to control from outside the relevant consciousness. Thus, instead of a single output depending subtly on many other variables, the output depends almost exclusively on a single input or small set of inputs.
I’d be happy to explain it in August, when I’ll be bored silly. At the moment, I’m pretty busy with my law school thesis, which is on antitrust law and has little to do with either free will or reductionism. Feel free to comment on any of my posts around that time, or to send your contact info to zelinsky a t gm ail dot com. Zelinsky is a rationalist friend of mine who agrees with you and only knows one person who thinks like me, so he’ll know who it’s for.
Thanks for bearing with me so far and for responding to arguments that must no doubt strike you as woefully unenlightened with a healthy measure of respect and patience. I really am done with both the free will discussion and the reductionist discussion for now, but I enjoyed discussing them with you, and consider it well worth the karma I ‘spent’. If you can think of any ways that what you see as my misunderstanding of free will or reductionism is likely to interfere with my attempts to help refine LW’s understanding of Goodhart’s Law, please let me know, and I’ll vote them up.
Why? How an algorithm feels is not a reliable indicator of its internal structure.
For convenience. If you show me a few examples where believing that I don’t have free will helps me get what I want, I might start caring about the actual structure of my mental algorithms as seen from the outside.
It is beneficial to believe you don’t have free will if you don’t have free will. From Surely You’re Joking, Mr. Feynman!:
All right, suppose all that is true, and that people can be hypnotized so that they literally can’t break away from the hypnotizing effect until released by the hypnotist.
That suggests that I should believe that hypnotism is dangerous. It would be useful to be aware of this danger so that I can avoid being manipulated by a malicious hypnotist, since it turns out that what appears to be parlor tricks are actually mind control. Great.
But, if I understand it correctly, which I’m not sure that I do, a world without free will is like a world where we are always hypnotized.
Once you’re under the hypnotist’s spell, it doesn’t do any good to realize that you have no free will. You’re still stuck. You will still get burned or embarrassed if the hypnotist wants to burn you.
So if I’m already under the “hypnotist’s” spell, in a Universe where the hypnotist is just an impersonal combination of an alien evolution process and preset physical constants, why would I want to know that? What good would the information do me?
I’m sorry, I’m not maintaining that free will is incompatible with determinism, only that sometimes free will is not present, even though it appears to be. When hypnotized, Richard Feynman did not have (or, possibly, had to a greatly reduced extent) free will in the sense that he had free will under normal circumstances—and yet subjectively he noticed no difference.
It appears to me that you created your bottom line from observing your subjective impression of free will. I suggest that you strike out the entire edifice you built on these data—it is built on sand, not stone.
I see; I did misunderstand, but I think I get your point now. You’re not claiming that if only Mr. Feynman had known about the limits of free will he could have avoided a burn; you’re saying that, like all good rationalists everywhere, I should only want to believe true things, and it is unlikely that “I have free will” is a true thing, because sometimes smart people think that and turn out to be wrong.
Well, OK, fair enough, but it turns out that I get a lot of utility out of believing that I have free will. I’m happy to set aside that belief if there’s some specific reason why the belief is likely to harm me or stop me from getting what I want. One of the things I want is to never believe a logically inconsistent set of facts, and one of the things I want is to never ignore the appropriately validated direct evidence of my senses. That’s still not enough, though, to get me to “don’t believe things that have a low Bayesian prior and little or no supporting evidence.” I don’t get any utility out of being a Bayesianist per se; worshipping Bayes is just a means to an end for me, and I can’t find the end when it comes to rejecting the hypothesis of free will.
Robin, I’ve liked your comments both on this thread and others that we’ve had, but I can’t afford to continue the discussion any time soon—I need to get back to my thesis, which is due in a couple of weeks. Feel free to get in the last word; I’ll read it and think about it, but I won’t respond.
Understood.
My last word, as you have been so generous as to give it to me, is that I actually do think you have free will. I believe you are wrong about what it is made of, just as the pre-classical Greeks were wrong about the shape of the Earth, but I don’t disagree that you have it.
Good luck on your thesis—I won’t distract you any more.
I place a very low probability on my having genuine ‘free will’ but I act as if I do because if I don’t it doesn’t matter what I do. It also seems to me that people who accept nihilism have life outcomes that I do not desire to share and so the expected utility of acting as if I have free will is high even absent my previous argument. It’s a bit of a Pascal’s Wager.
Why do you define “free will” to refer to something that does not exist, when the thing which does exist—will unconstrained by circumstance or compulsion—is useful to refer to? For one, its absence is one indicator of an invalid contract.
I’m not exactly sure what you’re accusing me of. I think Freedom Evolves is about the best exposition of how I conceive of free will. I am also a libertarian. I find it personally useful to believe in free will irrespective of arguments about determinism and I think we should have political systems that assume free will. I still have some mental gymnastics to perform to reconcile a deterministic material universe with my own personal intuitive conception of free will but I don’t think that really matters.
I’m confused. I haven’t read Freedom Evolves but Dennet is a compatiblist, afaik.
I think you’re saying you’re a compatibilist but act as if libertarianism were true, but I’m not sure.
I don’t really understand what you mean when you use the word ‘libertarian’ - it doesn’t seem particularly related to my understanding. I mean it in the political sense. Perhaps there is a philosophical sense that you are using?
Libertarian is the name for someone who believes free will exists and that free will is incompatible with determinism. Lol, it didn’t even occur to me you could be talking about politics.
I swear, if there ever exists a Less Wrong drinking game, “naming collision” would be at least “finish the glass”.
Ok, I’ve done some googling and think I understand what you meant when you used the word. I’d never heard it in that context before. I guess philosophically I’m something like a compatibilist then, but I’m more of an ’it’s largely irrelevant’ist.
I see. The word “genuine” is important, then—a nod to the “wretched subterfuge” attitude toward compatibilist free will. I withdraw my implications.
(I read Elbow Room, myself.)
No! A world without libertarian free will is a world exactly like this one.
ETA: Robin’s point, I gather, is that a world without libertarian free will is a world where hypnotism is possible. Which, as it turns out, is this world.
I was actually making a lesser point: that the introspective appearance of free will is not even a reliable indicator of the presence of free will, much less a reliable guide to the nature of free will.
Edit: From which your interpretation follows, I suppose.
(An aside: this sort of view is common in the free will dialectic—it is an incompatibilist theory, probably of the non-causal type. Jack’s objection is a standard one, and for good reason, I believe.)
Supernatural free will doesn’t exist. If your concept of free will requires it to be supernatural (mine doesn’t) and you are sure you experience free will clearly there is something wrong with either your concept or your perception.
Watch this: The Relativity of Wrong—http://www.youtube.com/watch?v=2tcOi9a3-B0
First of all, you cannot think when and about what you want. Same as you cannot want what you want. This leads to an infinite recursion. No system can understand itself for that the very understanding would evade itself forever. A bin trying to contain itself.
Anyway, there is no reason to go beyond physical, factual inquiry right now. You can assess your data with practicability. If a drug makes you think that you can fly you can jump from the next bridge and be brought back down to earth by reality.
Reality is not subject to interpretation, only description and abstraction. The fundamental nature of reality, its characteristics and qualities are absolute. Red is always red even when you call it green.
You don’t have to have a comprehensive grasp to determine the simplest of all conclusions: Something exists or it doesn’t. Either something is tangible or it doesn’t exist. My definition of subsistence is for something to have an influence on me. This also implies that something that exists can be subject to scientific inquiry. Something that exists can be assessed with practicability. It makes a difference.
If I cut my throat I may discover that I was dreaming or that I have been playing some advanced virtual reality game all along. Everything is possible. But right now there are safer and more promising options of gaining knowledge. How can I be sure? I can’t, but there is evidence which proved to be reliable so far. I have to suspect that it will continue to be reliable based on experiment and observation. That doesn’t make it the ultimate way of knowing or even a superior way but the best I know of at this time. And until I hit some hard barrier I do not have any good reason to try something else.