What is Probability?
January 21, 2017 9:41 PM   Subscribe

The Trouble with Quantum Mechanics - "The introduction of probability into the principles of physics was disturbing to past physicists, but the trouble with quantum mechanics is not that it involves probabilities. We can live with that. The trouble is that in quantum mechanics the way that wave functions change with time is governed by an equation, the Schrödinger equation, that does not involve probabilities. It is just as deterministic as Newton's equations of motion and gravitation. That is, given the wave function at any moment, the Schrödinger equation will tell you precisely what the wave function will be at any future time. There is not even the possibility of chaos, the extreme sensitivity to initial conditions that is possible in Newtonian mechanics. So if we regard the whole process of measurement as being governed by the equations of quantum mechanics, and these equations are perfectly deterministic, how do probabilities get into quantum mechanics?" (via)
The class of Lindblad equations contains the Schrödinger equation of ordinary quantum mechanics as a special case, but in general these equations involve a variety of new quantities that represent a departure from quantum mechanics. These are quantities whose details of course we now don't know. Though it has been scarcely noticed outside the theoretical community, there already is a line of interesting papers, going back to an influential 1986 article by Gian Carlo Ghirardi, Alberto Rimini, and Tullio Weber at Trieste, that use the Lindblad equations to generalize quantum mechanics in various ways.
also btw...
  • What is Probability? - "Probabilities may be subjective or objective; we are concerned with both kinds of probability, and the relationship between them."
  • [On] the origin of probability in quantum mechanics - "I give a brief introduction to many worlds or 'no wavefunction collapse' quantum mechanics, suitable for non-specialists. I then discuss the origin of probability in such formulations, distinguishing between objective and subjective notions of probability."
  • Decoherence does not resolve the collapse question, contrary to what many physicists think. Rather, it illuminates the process of measurement and reveals that pure Schrodinger evolution (without collapse) can produce the quantum phenomena we observe. This of course raises the question: do we need collapse? If the conventional interpretation was always ill-defined (again, see Bell for an honest appraisal [1]; Everett referred to it as a "philosophical monstrosity''), why not remove the collapse or von Neumann projection postulates entirely from quantum mechanics?

    The origin of probability is the real difficulty within many worlds interpretations. The problem is subtle and experts are divided as to whether it has been resolved satisfactorily. Because the wave function evolves entirely deterministically in many worlds, all probabilities are necessarily subjective and the interpretation does not require true randomness, thereby preserving Einstein's requirement that outcomes have causes.
  • 50 years of Many Worlds (Some recommended reading) - "Personally, I prefer to call it No Collapse instead of Many Worlds -- why not emphasize the advantageous rather than the confusing part of the interpretation?"
  • Feynman and Everett (Quantum correspondence) - "I believe that Zeh would agree with me that decoherence is merely the mechanism by which the different Everett worlds lose contact with each other! (And, clearly, this was already understood by Everett to some degree.) Incidentally, if you read the whole paper you can see how confused people -- including Feynman -- were about the nature of irreversibility, and the difference between effective (statistical) irreversibility and true (quantum) irreversibility."
  • Let's suppose you live in a deterministic world and are about to flip a coin. You assign a probability to the outcome because you don't know what it will be. In secret, the outcome is already determined. To you, the process appears probabilistic, but really it is not. That is actually how MW works, but this is not widely appreciated.
  • Is the wavefunction real? - "Quantum theorem shakes foundations: The wavefunction is a real physical object after all, say researchers."
  • The measure problem in many worlds quantum mechanics - "We explain the measure problem (cf. origin of the Born probability rule) in no-collapse quantum mechanics. Everett defined maverick branches of the state vector as those on which the usual Born probability rule fails to hold -- these branches exhibit highly improbable behaviors, including possibly the breakdown of decoherence or even the absence of an emergent semi-classical reality. An ab initio probability measure is necessary to explain why we do not occupy a maverick branch. Derivations of the Born rule which originate in decision theory or subjective probability do not resolve this problem, because they are circular: they assume, a priori, that we reside on a non-maverick branch."
  • To put it very succinctly: subjective probability or decision theoretic arguments can justify the Born rule to someone living on a non-maverick branch. But they don't explain why that someone isn't on a maverick branch in the first place.
    It seems to me absurd that many tens of thousands of papers have been written about the hierarchy problem in particle physics, but only a small number of theorists realize we don't have a proper (logically complete) quantum theory at the fundamental level."
  • Weinberg on quantum foundations - "Considerable progress has been made in recent years toward the resolution of the problem, which I cannot go into here."
oh and :P
When They Came from Another World - "What if the future is as real as the past? Physicists have been suggesting as much since Einstein. It's all just the space-time continuum. 'So in the future, the sister of the past', thinks young Stephen Dedalus in Ulysses, 'I may see myself as I sit here now but by reflection from that which then I shall be'. Twisty! What if you received knowledge of your own tragic future—as a gift, or perhaps a curse? What if your all-too-vivid sensation of free will is merely an illusion?"
posted by kliuless (68 comments total) 93 users marked this as a favorite
 
i feel a strong emotion inclination to believe probability is an aliasing artefact of the simulator
posted by compound eye at 12:04 AM on January 22, 2017 [15 favorites]


I taught a module on free will last semester, and since everyone always bleats 'but what about quantum mechanics?' (although it almost certainly makes no difference to human actions) I thought I'd be conscientious and get to the bottom of the arguments against hidden deterministic variables. Since I'm no expert in physics, I also got a professor in the physics dept to check over my slides.

I was amazed to find that the deterministic Bohmian/pilot wave interpretation can account for all the same observations as the indeterministic Copenhagen interpretation. That indeed, both theories assume 'spooky' action at a distance, but only Copenhagen includes the extra non-realist assumption (that reality is fundamentally probabilistic). The only advantage of Copenhagen seems to be that physicists have reconciled it with special relativity- but as far as I can tell, there's no principled reason why Bohmian mechanics couldn't also do this.

Since there's no experimental observation that distinguishes Copenhagen and Bohm (or many worlds) we should either suspend judgement- or for purely philosophical reasons prefer Bohm because it's the more parsimonious theory and coheres with the principle that nothing just randomly happens (i.e. the principle of sufficient reason). That's my Leibnizian take on the matter.
posted by leibniz at 12:58 AM on January 22, 2017 [12 favorites]


"everything is possible, fewer things are probable."

(overheard recently)
posted by philip-random at 1:16 AM on January 22, 2017 [5 favorites]


I wouldn't say Bohm is the more parsimonious theory, as it includes both a pilot wave AND a particle. Reconciling QM with special relativity is not a minor detail. It's a prerequisite for making sense of decades of theoretical and experimental work.
posted by Schmucko at 2:04 AM on January 22, 2017 [7 favorites]


I was amazed to find that the deterministic Bohmian/pilot wave interpretation can account for all the same observations as the indeterministic Copenhagen interpretation.

Oh man, do I have an experiment to show you :)

Short version: it was noticed some time ago that if you put a dish of light oil (silicone is good) over a speaker and drive the speaker at approx 80 Hz you can take a drop of the same oil and it will bounce perpetually on the surface of the rest of the oil, never quite managing to merge with it, whilst creating a standing wave in the oil around it. This pair - the wave & the droplet can shift to a 'walking regime' where they form a moving pair - the droplet skips across the surface of the oil, creating the wave that pushes it along. (The energy to drive this & overcome air resistance is coming from the vibration from the speaker.)

About a decade ago, Yves Couder and Emmanuel Fort investigated the mathematics of this phenomenon & it turned out that the wave on the surface of the oil and the droplet interact in exactly the same way as a particle & it’s associated pilot wave in Bohm-style quantum mechanics.

You can probably see where this is going already, but you should definitely watch this video from the Veritasium series on YouTube, and then follow up with the researcher’s website. More theory here.
posted by pharm at 2:08 AM on January 22, 2017 [24 favorites]


Coincidentally just came across this blog post critical of the Bohmian interpretation.
posted by Schmucko at 2:46 AM on January 22, 2017 [1 favorite]




Humans seem to have a problem with free will. They don't like living in a universe without it.

Well, bad news people. There is no model of the universe that includes free will.

If you start by accepting the Many Worlds interpretation (which is right there in the mathematics jumping up and down and shouting) and you leave free will out of the model (because it is simply not there in any model) then it all makes (some sort of) sense.

We don't have free will and the universe cascades down all the pathways.

Pass me another beer.
posted by Combat Wombat at 6:53 AM on January 22, 2017 [6 favorites]


There is no model of the universe that includes free will.

Sure there is. Or at least allows for the possibility of free will. We can never actually know any more than we can ever observe the universe through any other lens than our own individual consciousnesses.

Which is itself part of the biocentric view that everything we observe, reality itself, our brains themselves, and our perception of our own minds, is generated by consciousness. In this way of thinking, all of this quantum hoodooery are artifacts of the mechanism (whatever THAT is) of our consciousness.

Here, many are inclined to assume that there is a reality external to our minds, but this, by definition, cannot be known. Likewise, it cannot be known whether we have free will, but hey, that's still a maybe!

Note that this is not necessarily my contention for the One True Reality™, but there are serious people with good reasons to consider this.
posted by cmoj at 7:49 AM on January 22, 2017 [3 favorites]


There are only two situations that make sense. Either the universe is non-deterministic (i.e., probabilistic), or the universe is deterministic, but completely unpredictable. Note that this is a distinction without a difference, as if there is no way to predict the future, then how do you test it for determinism?. A lot of it boils down to entropy and causality. If the universe was deterministic and predictable, It would seem to violate the laws of thermodynamics, and we'd get perpetual energy to come along for the ride. Also, time (specifically, the passage of time) doesn't really make a lot of sense in a deterministic universe. In a perfectly symmetrical universe, why would we experience one dimension in a fundamentally different way than the others?
Also, why would there be an abundance of matter over anti-matter from the big bang if the universe was symmetrical? And if it's possible for a deterministic universe to be asymmetrical, why and what would cause that asymmetry?

To me the simpler answer is actually we live in a random universe, at lest at the quantum level.

The Many Worlds theory at first glance appears to have a cozy explanation that bridges determinism and randomness. But all it really does is shift the question from "what is the future" to "which of the infinite future paths gets chosen". Or to put it another way, where do those other worlds go once the future has become the present?

So yeah, I think free will is possible, maybe even probable. Nudge the universe a little bit differently and the outcome changes, and you can't go back.
posted by forforf at 8:02 AM on January 22, 2017 [1 favorite]


There is no model of the universe that includes free will.

This depends, of course, on how you define "model". In the sense of having an equation or physical model which has room for a "free will" variable, I believe the statement is correct. I would venture that most people have a "model" of the universe that includes free will, but these models are somewhat less rigorously-specified than what would be accepted by a practitioner.
posted by Slothrup at 8:03 AM on January 22, 2017


Obviously, probabilities must all be positive numbers, and add up to 100 percent.

I'm probably an idiot, but particularly in a many-worlds interpretation of QM, I don't see anything fundamental, basic, or obvious about this.
posted by Western Infidels at 8:23 AM on January 22, 2017 [2 favorites]


If free will exists and we act like it exists, we can make things better for ourselves.
If free will exists and we act like it doesn't exist, we allow the world to do whatever it wants to us, which will largely not be to our benefit.
If free will doesn't exist, the world does whatever it will do to us no matter how we think or act (although the words "think" and "act" aren't actually meaningful in this interpretation anyway), which will largely not be to our benefit.

There's only one way that makes sense for an intelligent person to think or behave on this matter, it turns out. Behaving as though free will does not exist never leads to better outcomes than behaving as though it does. Although I guess acting really jaded about how free will can't exist might impress some teenagers or something, which depending on your goals can have some local benefit.

(Also, a description of the many-worlds interpretation as "right there in the mathematics jumping up and down and shouting" is pretty strange, given that it assumes several things that are not yet in evidence and isn't really capable of describing how reality moves from one state to another. It's an interesting idea, but as a theory it is rather incomplete.)
posted by IAmUnaware at 8:25 AM on January 22, 2017 [9 favorites]


I remember this building maintenance guy who often used to ride the bus going home from university. He was a big stout Fred Flintstone kind of guy, never took off his yellow hard hat, and was often drunk after work. Once he sat beside me and ranted the whole way about how everything was predestined. I decided to get off the bus a couple of stops early and walk the rest of the way.
posted by sneebler at 8:49 AM on January 22, 2017 [4 favorites]


If free will exists and we act like it doesn't exist, we allow the world to do whatever it wants to us, which will largely not be to our benefit.

I really wanted to resist posting in this thread and yet...

I don't think the term "free will" in this context really means quite what people generally mean by it. I like to think I choose to do what's best for us all and I like to think that's my choice. I could prove that, by choosing to behave differently, but I don't want to so I won't. Whether or not I've made that choice freely is not really something I can properly discover. But if it's determined by some external force then I can't detect it so the question is moot from my point of view. So yes, I tend to agree that...

There's only one way that makes sense for an intelligent person to think or behave on this matter
posted by merlynkline at 9:17 AM on January 22, 2017 [3 favorites]


No principle of freedom can be discovered by an examination of matter. Freedom is only found when we acknowledge consciousness as omnipresent, infinite and eternal. Physicalism and freedom are mutually exclusive.
posted by No Robots at 9:42 AM on January 22, 2017


There is no model of the universe that includes free will.

The model of the universe used to exclude things happening faster than light but Alain Aspect certainly put a bullet in that limitation.
posted by Talez at 10:03 AM on January 22, 2017


You guys are missing the point. Completely. All you need to know is one simple fact.

It's turtles.

All the way down.
posted by Samizdata at 10:20 AM on January 22, 2017 [8 favorites]


That's only because people back then didn't know of the Technodrome.
posted by Talez at 11:11 AM on January 22, 2017


> everyone always bleats 'but what about quantum mechanics?'

That's a pretty shitty way to think about your students. Would you prefer they just silently absorb your pearls of wisdom?

> There is no model of the universe that includes free will.

This is so obviously wrong it's a clear cover for your own discomfort with the idea of free will (which, to be fair, you share with many worshipers of Almighty Science).
posted by languagehat at 11:16 AM on January 22, 2017 [5 favorites]


There is no model of the universe that includes free will.

I would say this is triviality true, because there is no coherent model of "free will" to begin with.

The assumption seems to be that it involves some non-causal phenomenon, that there is some sort of numinous "conscious will" exerting its influence on the physical universe from "outside".

I've read a lot of fancy language dressing up this concept, but nothing to convince me that there is really any "there" there.

As others have pointed out, "free will" is a concept that makes sense in the philosophy of human affairs. This need to prove it in terms of fundamental physics strikes me as desperate and myopic.
posted by bjrubble at 11:32 AM on January 22, 2017 [6 favorites]


The assumption seems to be that it involves some non-causal phenomenon, that there is some sort of numinous "conscious will" exerting its influence on the physical universe from "outside".

Perhaps this conscious will is not outside, but inside, ie. inherent in all phenomena.

As others have pointed out, "free will" is a concept that makes sense in the philosophy of human affairs. This need to prove it in terms of fundamental physics strikes me as desperate and myopic.

A philosophy that doesn't account for all phenomena isn't really a philosophy.
posted by No Robots at 11:52 AM on January 22, 2017 [1 favorite]


Well, John Horton Conway, who has thought quite deeply at least about the foundations of mathematics, has proved what he calls a "Free Will Theorem" about quantum mechanics. Personally I think that some connotations of the classical idea of free will are a bit inconsistent. However, I have no problem with the idea that some things in our universe happen randomly without anything to point to a deterministic cause. I think many Bohm model adherents are stubbornly refusing the simplest and most powerful ways of expressing and using QM for the sake of their mental inflexibility.
posted by Schmucko at 11:54 AM on January 22, 2017


The assumption seems to be that it involves some non-causal phenomenon, that there is some sort of numinous "conscious will" exerting its influence on the physical universe from "outside".

Very much so. I think a critical flaw in the discussions of free will (at least those I have seen) is that no one has carefully defined the subject to which the free will is being ascribed. First, define the subject (and none of this Cartesian duality crap). Then, draw a boundary around the subject. Inside there the free will happens, and you don't get to look inside. That free will acts on the not-subject (the rest of the universe).

Perhaps this conscious will is not outside, but inside, ie. inherent in all phenomena.

Considering that subjects that we consider conscious (e.g., individual people) are instances of the same phenomena that are the all phenomena everywhere, I think this has to be true, or nothing is conscious at all. There doesn't have to be any woo here: self-reflexive consciousness does require a feedback loop, and memory requires some writeable and readable storage medium, but it's all made of the same stuff and following the same natural laws as everything everywhere.
posted by heatherlogan at 12:36 PM on January 22, 2017


on the free will, quantum whatever tip ...

Many years ago, I attended a workshop given by Robert Anton Wilson. I remember during one of the breaks listening in as he and one of the other workshoppers discussed the free will question. As she put it, "Sometimes I'm a mother, other times I'm at work which makes me a teacher, other times I'm in a movie theater and thus an audience member, or shopping which makes me a consumer." My objective role keeps changing but it's still always me. Is this not the same thing with light being sometimes being a wave form, other times a particle, and does this not suggest that whatever IT it is, this phenomena that we think of as "light", it's constantly adjusting to the given situation? In other words, light itself seems to have a will. In other words, the Universe ..."

Wilson smiled and nodded.
posted by philip-random at 1:20 PM on January 22, 2017 [1 favorite]


How do stochastic patterns enter deterministic systems? Through poorly designed statistics. Ask an ecologist, or, more precisely, an ecological analyst, they ve had to deal with this problem for 40 years.
posted by eustatic at 1:47 PM on January 22, 2017


Personally I think that some connotations of the classical idea of free will are a bit inconsistent. However, I have no problem with the idea that some things in our universe happen randomly without anything to point to a deterministic cause.

My issue is that I've never heard a convincing argument for why these two things (free will and true randomness) would have anything to do with each other. The arguments I have heard (imo) boil down to "the randomness is what allows free will to exert itself on the physical universe" -- which to me equates to "you can't disprove dualism."

The Flying Spaghetti Monster is a dualist, by the way. Well known fact.
posted by bjrubble at 2:07 PM on January 22, 2017 [4 favorites]


If it's free, it's not willed; if it's willed, it's not free.
posted by Wataki at 3:36 PM on January 22, 2017 [2 favorites]


I don't know if I have free will, but from my point of view, all of you out there certainly do.
posted by tgyg at 7:43 PM on January 22, 2017 [1 favorite]


My issue is that I've never heard a convincing argument for why these two things (free will and true randomness) would have anything to do with each other.

I used to think that they had absolutely nothing to do with each other. And I still think that free will is compatible with determinism. But I had a glimmer recently of how one might think that free will and objective chance are connected. Here's a stab at making the connection clear. (On preview, this is longer than I intended. Apologies for the length.)

Consider a case where you do something. Maybe this is moving your hand or telling a joke or whistling. Call the thing you do X. The question of free will is essentially the following: If you had tried to do Y instead of X (where Y is a distinct action), would you have done Y instead of X? A "yes" answer means you had free will, while a "no" answer means you didn't.

Ordinarily, we answer such questions "yes" quite frequently. I could have made a salad this evening, instead of red curry with rice. If I had tried to eat a salad, I would have succeeded. I have all the "fixings" in my fridge, I know how to shred lettuce and cut up tomatoes, carrots, radishes, and so on. No one would have tried to stop me. I simply didn't want to have a salad. I didn't try to make one. So, I didn't make one.

Notice that the line I'm taking here is couched in terms of abilities and capacities that I have. And this, too, is completely ordinary. I have the ability to play the trombone (though it's been a long time). That doesn't mean I am now exercising that ability. In general, we don't have to be actually exercising an ability in order to have that ability.

Anyway, what one could do if one had tried, figuring out what abilities one has, or figuring out what capacities a thing has, are all modal issues. Unlike claims like, "There is a table in front of me," it is not obvious how to determine the truth value of claims like, "If I had knocked over the table, it would have made a mess," by looking at the actual state of the world. Some reasoning is required in order to evaluate modal claims. Lots of suggestions have been made for how to do this. Here is one suggestion (made by philosopher of physics, Tim Maudlin) for how to evaluate counterfactual conditionals (conditionals like, "If I had tried, I would have made a salad," where the antecedent is known to be false) that is salient for the current discussion: Take a space-like surface in the spacetime where you live; set all of the values on the surface such that the antecedent of the counterfactual comes out true; then plug that surface into the physical laws and roll it forward in time to see whether the consequent comes out true.

So, now consider my possible salad again. I say that if I had tried, I would have made a salad. To test this according to Maudlin's suggestion, I need to go back to take a point in time earlier in the evening and set all of the physical values such that "I try to make a salad" comes out true. Then I roll the laws forward to see whether "I make a salad" comes out true later on.

So far, all of this could be used to illustrate how free will is compatible with determinism. Even if the laws are deterministic, if I evaluate counterfactuals in the way that Maudlin suggests, it will often be the case that if I had tried to do something other than what I did, then I would have done something different than what I actually did. It is at this point that things get interesting. One might say, "Wait, that's not what I mean when I talk about free will! You're changing the way the world is in evaluating your counterfactual -- making it so that you tried to do something that you actually didn't try to do at all! What I want to know is whether you could have made a salad leaving fixed all of the earlier facts about the universe." Now, this demand is not especially clear. If the demand is that the entire universe remain fixed, while it remains open that we could do something other than what we did, then it is obvious that we don't have free will. You don't need any special physical theory, you don't need determinism, you don't even need materialism (or physicalism or whatever you want to call it). All you need is classical logic. So, let's relax the demand a bit and say that we want to leave everything in the universe fixed right up until the moment of the event. In that case, proponents of free will should say that the only relevant question is whether I am ever free to try to do anything other than what I in fact try to do. And here is where objective chance might come into the story.

Suppose I have an objectively-chancy coin. (We know that real coins are not really chancy. But it's a useful heuristic. If you want something more plausibly objectively chancy, take a nuclear decay process as the example.) The coin will come down either heads or tails and has an objective chance of 1/2 of coming down heads. One day, I play a game with a friend. If the coin comes up heads, I pay my friend a dollar. If it comes up tails, he pays me a dollar. (We're bored, what can I say?) I flip the coin, and it comes up tails. What does Maudlin's suggestion say about how to evaluate the claim (which looks entirely sensible) that the coin could have come up heads? Maudlin says to take the surface that actually obtained right before the target event. Roll the laws forward and see what happens. In this case, since the coin is objectively chancy, when you roll the laws forward, there are two models (in the sense of model theory in logic or in the sense of the semantic view of scientific theories) for the law. One model has the universe exactly as the actual universe. The other model has the universe with a heads-up coin and whatever else has to be different in order to accommodate that fact.

Long story shorter: If there is objective chance and if trying is an objectively chancy thing, then quite sensible theories of modality say that in some cases, you could have tried to act in a way other than the way you actually acted, even if the entire history of the universe up until your action remains fixed.

This is not a story according to which randomness allows free will to exert itself on the physical universe. In the account I've just sketched, free will isn't a thing. Rather, the account says that an agent is free -- acts freely, has a free will -- if the agent has the ability to try to act in various ways and if by trying to act in some specific way, the agent would succeed in doing what she tries to do. Moreover, there is no dualism in the story. Everything I've said here is completely consistent with physicalism.
posted by Jonathan Livengood at 8:49 PM on January 22, 2017 [6 favorites]


Well, bad news people. There is no model of the universe that includes free will.

That's not true. I still maintain it's not even a scientific question in the sense people make out it to be. The Compatibilist argument still wins the day for me: there's got to be something internal to the state of people-objects (to use object oriented analysis concepts from comp sci) that determines how a decision goes. The same person doesn't necessarily make the same choice as someone else or even consistently in the same situation. Some small part of how a person-object behaves over time--the specific curve and shape of their time line--is uniquely defined by properties internal to the person, that form the person's character, which is itself formed by outside factors (genetics, life experience, social context), but once formed, allows for limited agency and the ability to generate non-deterministic behaviors (to act differently in identical external circumstances) in the same way meaningful words are built up out of individually meaningless sounds. The thing that confuses the question is refusing to stop at the boundary of the self when analyzing causality. Sure, if you have God-like knowledge about the inner workings of a person, you can predict perfectly how they'll behave. No one would argue that means they have no agency.
posted by saulgoodman at 4:21 AM on January 23, 2017 [3 favorites]


Long story shorter: If there is objective chance and if trying is an objectively chancy thing, then quite sensible theories of modality say that in some cases, you could have tried to act in a way other than the way you actually acted, even if the entire history of the universe up until your action remains fixed.

But "chance" here -- to the extent that it is analogous to an objectively random coin flip -- is meaningless. How is following a coin flip any more a "choice" than following a set of fixed rules?

Sure, if you have God-like knowledge about the inner workings of a person, you can predict perfectly how they'll behave. No one would argue that means they have no agency.

I think this is exactly what a lot of people are arguing.
posted by bjrubble at 6:20 AM on January 23, 2017 [1 favorite]


The more complicated the machine, the easier it is to anthropomorphize it.
posted by Pope Guilty at 6:28 AM on January 23, 2017 [1 favorite]


Sure, if you have God-like knowledge about the inner workings of a person, you can predict perfectly how they'll behave. No one would argue that means they have no agency.

I think this is exactly what a lot of people are arguing.


And we come full circle, because quantum mechanics precludes any such god-like knowledge about the inner workings of a person. That's a Newtonian conception, now known to be false.
posted by heatherlogan at 6:58 AM on January 23, 2017 [2 favorites]


Thanks for this post, I hadn't read much about pilot wave theories before, and it's pretty great. The oil drops analogue is particularly fascinating.
posted by vibratory manner of working at 7:37 AM on January 23, 2017


Great post! Here is a question about many worlds: if a small lump of uranium atoms is watched carefully, after a certain amount of time half will have decayed. An individual atom decays at any given moment with probability x. In a rough picture of the many worlds interpretation, that atom decays at every moment, but only in a fraction of the branches of history. It seems as though this implies that in some history, all of my uranium atoms decay at once! Of course this would be very,very rare, and it would be very odd to find yourself in such a world, right?
posted by TreeRooster at 8:36 AM on January 23, 2017


Sure, if you have God-like knowledge about the inner workings of a person, you can predict perfectly how they'll behave. No one would argue that means they have no agency.

as a some time writer of fiction, I may not have God-like knowledge of any particular person, but I do of my characters. And trust me when I say, I really can't predict how they'll behave. They surprise me all the time. They have agency.
posted by philip-random at 10:39 AM on January 23, 2017


Correct me if I'm wrong, but Bohmian pilot-wave quantum theory is non-relativistic, analogous to the non-relavistic Schödinger equation. There's no equivalent approach to relativistic quantum mechanics, more commonly known as quantum field theory.
posted by fantabulous timewaster at 11:55 AM on January 23, 2017 [1 favorite]


But "chance" here -- to the extent that it is analogous to an objectively random coin flip -- is meaningless. How is following a coin flip any more a "choice" than following a set of fixed rules?

On the first bit ... it sure seems meaningful. In fact, I think I can give a perfectly good technical account. An event is objectively chancy iff there is no unique trajectory in the forward direction according to the physical laws (or theories).

As to the second, I never said anything about choice. All I was doing was making sense of the idea that (at least for some beings) on occasions when one actually tries to do X, one could have tried to do something other than X, even if we leave the entire past history of the universe fixed. It's the bit about leaving the whole past history fixed that makes objective chance (potentially) relevant to the question of free will.

Maybe you are worried that if trying is objectively chancy, then we don't have any genuine ownership of what we try to do? If that is your worry, I'm not sure why I should share it. Could you say more? I think I just want to say, "This is what I tried to do, but I could have tried to do something else." When pressed to say what I mean by that, I give a story similar to the one I sketched earlier. When someone objects that I don't have free will in that case, I reply, "This is all that anyone has ever wanted or plausibly could demand from free will: that we could have done otherwise than we actually did. And the account I've given shows how to make sense of such claims."
posted by Jonathan Livengood at 12:32 PM on January 23, 2017 [3 favorites]


And we come full circle, because quantum mechanics precludes any such god-like knowledge about the inner workings of a person. That's a Newtonian conception, now known to be false.

But why is the truth or falsity of this relevant? QM implies that such knowledge is impossible, because there is an element of true randomness. But such knowledge is practically impossible even in a Newtonian model, so what is it in the nature of "free will" that it can exist in the one model but not in the other?

The implication seems to be that adding a dice roller to an otherwise deterministic machine can somehow imbue it with "free will." The only way this makes any sense to me is if it is really just dualism in another guise -- that if pushed, the claim is actually that the QM "randomness" in the brain isn't actually random but carries the "signal" from some nonphysical mind.
posted by bjrubble at 4:09 PM on January 23, 2017 [1 favorite]


Maybe you are worried that if trying is objectively chancy, then we don't have any genuine ownership of what we try to do?

I'm not worried at all. I regard "free will" as a useful concept in evaluating my own behavior and that of others. I see no need for it to map onto some intrinsic, low-level property of the universe.

My choices are predictable, in that they are rational choices based on my perceptions and desires, which are themselves the logical product of the particular historical contingency that is "me." The idea that someone with impossibly comprehensive knowledge about me could predict my choices doesn't make them any less "my" choices. Nor does rolling a die solely to fluster this theoretical omniscient predictor make them any more "my" choices. (Quite the opposite, in fact.)

I think people are tying themselves into crazy (and ineffective) knots trying to square a circle that can't (and really doesn't need to be) squared.
posted by bjrubble at 4:37 PM on January 23, 2017 [2 favorites]


bjrubble: the brain is a part of what makes a self. From my POV, it's a category error to discuss free will at a level of abstraction below that which recognizes the self as an atomic unit in the analysis. People have claimed selves have something called free will. Selves are not just brains, but a whole network of things, including social components. But once you've started deconstructing the self as a first step in the analysis, of course that analysis won't ever find evidence of free will, because the method of analysis already takes as axiomatic the idea there's no atomic self in reality by deconstructing the self down to its parts. Just as you wouldn't be able to "ride" a bicycle that's been stripped to parts jumbled up in a big sack, you can't expect an analysis that operates at a lower level of abstraction than a high level idea like "self" to produce results that are meaningful at that higher level of abstraction. Selves don't exist at a low level of abstraction, but that doesn't mean they don't exist absolutely anymore than the fact you can take a bike apart and jumble the parts up in a bag means bikes don't exist. It misses the point to ever argue about what happens inside brains when discussing free will. Whatever happens in the brain is only a part of the self, not the whole. Isn't it the compositional fallacy to look for evidence of some capability or property at the level of the parts when the capability or property only exists at an emergent level? Nobody would argue the E major chord doesn't exist or is really just the same thing as the note E. Free will/selves/bicycycles/other higher order concepts like that have to be understood and analyzed at the same level of abstraction on which they function to make any sense.
posted by saulgoodman at 5:10 PM on January 23, 2017 [4 favorites]


(I guess that's actually the flipside of the compositional fallacy: pointing out that none of the parts that make up a thing have some particular quality or property ascribed to the thing as a whole proves nothing.)

Even if people really operated purely like machines (putting aside how arrogant that metaphor really is--machines being simplified imitations of useful subsets of features we've learned to mimic from more complex biological and natural systems, not the other way around), that couldn't possibly tell us anything at all about whether or not some form of agency exists at the level of the whole machine. All that's required to prove that is to demonstrate it's possible for a human to make different choices given basically identical external, initial conditions, and if that's not possible, there's no point in teaching anyone to read or to learn any critical thinking skills except that, for some reason, we're compelled to by forces beyond our control. All of which seems like nonsense to me.
posted by saulgoodman at 5:27 PM on January 23, 2017 [1 favorite]


From my POV, it's a category error to discuss free will at a level of abstraction below that which recognizes the self as an atomic unit in the analysis.

Yes, thank you! "Category error" is a much better description of what I was trying to say. Literally, "free will" as it applies to humans and "indeterminism" as it applies to raw physics are two completely different things in two completely different domains. They are merely vaguely analogous.
posted by bjrubble at 6:06 PM on January 23, 2017 [2 favorites]


TreeRooster:: if a small lump of uranium atoms is watched carefully, after a certain amount of time half will have decayed. An individual atom decays at any given moment with probability x. In a rough picture of the many worlds interpretation, that atom decays at every moment, but only in a fraction of the branches of history. It seems as though this implies that in some history, all of my uranium atoms decay at once! Of course this would be very,very rare, and it would be very odd to find yourself in such a world, right?

It would be very, very odd indeed. Roughly equivalent to finding yourself in the world where all the atoms of the air around you spontaneously end up in the corners of the room, leaving you gasping for breath for a few seconds.

One of the (many) criticisms of many worlds is that it seems to imply a ridiculous multiplicity of almost identical worlds which differ only in a photon here or there & whether a particular uranium atom decays now or later.

Of course, some of these decays have macroscopic effect: e.g. It’s *this* uranium decay which kicks out a gamma ray which corrupts the DNA of Bob so that he gets cancer when he’s 35. But the vast majority have no such macroscopic effect & creating a world for each seems like multiplying entities beyond all reason. You can, of course, assert that only quantum events that end up having some kind of macroscopic effect end up actually splitting the world, but since those macroscopic effects can be delayed by decades, it seems difficult to make this work in a satisfying way.
posted by pharm at 12:40 AM on January 24, 2017


All that's required to prove that is to demonstrate it's possible for a human to make different choices given basically identical external, initial conditions, and if that's not possible, there's no point in teaching anyone to read or to learn any critical thinking skills except that, for some reason, we're compelled to by forces beyond our control. All of which seems like nonsense to me.

If I understand bjrubble's point correctly, it's that a non-deterministic universe could satisfy this proof, with the "different choices" being possible but merely as uncaused coin-flip variations in behavior (but independent of the identical external initial conditions, as you require); hence the ability to choose is lacking the visceral meaningfulness and "point" of being able to make those "choices" because the aspect of them that's independent of external initial conditions is only an (objectively-chancy) coin flip rather than anything integrally connected to the nature of the self and identity.
posted by XMLicious at 3:30 AM on January 24, 2017


Remember folks if it's free, that just means you're the product being sold!
posted by cmoj at 11:54 AM on January 24, 2017 [1 favorite]


XMLicious: Yes, exactly. Randomness is inherently meaningless, by definition it conveys no intent or purpose, so it seems bizarre to hold that up as the essential core of free will.

Imagine someone who's known you your whole life, has been with you through everything, and you share a passion for, say, giant robot mecha. And they see this awesome video that they know is right up your alley, and show it to you.

Does the fact that they knew you would like it make your liking of it any less "authentic"? If you reran the tape, and sometimes you liked it even more, or a bit less -- not because your friend didn't really know you so well, but actually for no reason at all -- would that give you more confidence in your "ownership" of your reaction?

Deep down, the universe is pretty clearly a fairly simple rule-based affair. And yet it can give rise to this amazing phenomenon that thinks, and understands, and embodies concepts like "meaning" and "self." How this happens is in large part up for grabs.

But one thing is clear: deep down, the universe is just some fairly simple rules. Really, no two ways about it. Magnets attract, mass produces gravity, and so on.

Now, these simple rules -- which, again, by pretty much any account aren't actually where the magic happens -- why does it matter whether they involve a "real" random number generator?
posted by bjrubble at 9:40 PM on January 24, 2017 [2 favorites]


But such knowledge is practically impossible even in a Newtonian model, so what is it in the nature of "free will" that it can exist in the one model but not in the other?

The main issue here is that whether a law of nature is deterministic is not equivalent to whether or not we can predict things on the basis of the law. At a first pass, the laws of nature are deterministic if the laws plus the past history of the universe entails a unique future development of the universe. Many people have thought that Newton's theory is deterministic in this sense. (Some philosophers of my acquaintance have argued that it is not, but that is controversial (pdf). But for the present, let's assume that Newton's theory is deterministic.) But Newton's laws might be deterministic in that sense without allowing for prediction. If the laws of nature are genuinely chancy (in the sense that they are not deterministic), then the laws plus the past history of the universe do not entail a unique future development of the universe.

Perhaps genuine chanciness in this sense is not required for free will. Perhaps we can have free will even in a universe that only develops in one way given any initial part of its history together with its laws. That view is compatibilism about determinism and free will, and I tend to think that compatibilism is the correct view. Even in a deterministic universe, it will often be the case that if I had tried to do something other than what I actually did, I would have succeeded in doing that thing. But. Some people think that the truth of that counterfactual is not enough. They think that it also has to be the case that I could have tried to do something other than I actually tried to do. Now, one could simply iterate here and say, "Well, if I had tried to try to do something other than what I actually tried to do, then I would have succeeded in trying to do that other thing." But it seems that to satisfy this new demand -- that it be true that in those very circumstances I could have tried to do something other than what I actually did -- we would have to iterate indefinitely, and that doesn't look like a winning strategy for defending the claim that we have free will.

Anyway, the view I suggested shows how it could be true that in those very circumstances I could have tried to do something other than what I actually tried to do, provided the universe has some genuine chanciness in it.

The implication seems to be that adding a dice roller to an otherwise deterministic machine can somehow imbue it with "free will." The only way this makes any sense to me is if it is really just dualism in another guise -- that if pushed, the claim is actually that the QM "randomness" in the brain isn't actually random but carries the "signal" from some nonphysical mind.

Not at all. You have the metaphysical picture exactly backwards here. On the picture being sketched, the chanciness of my various tryings partially constitutes the being that I am. It isn't that there is a ghostly soul that sends signals to my physical body by way of a genuinely noisy comm channel. Rather, it is the fact that I am the sort of physical being that has some chancy parts that makes the label "free" appropriate. At least, this (or something nearby) is what I think a libertarian about free will ought to say.
posted by Jonathan Livengood at 1:36 AM on January 25, 2017 [1 favorite]


The way you're using the word "tried" is what's dissonant to me, and to bjrubble I think; even in the description of the self as an integrated whole to which the property of chanciness applies in entirety, why associate the chanciness with trying or willing or intention?

If it's the case that
Even in a deterministic universe, it will often be the case that if I had tried to do something other than what I actually did, I would have succeeded in doing that thing.
how does adding uncaused, non-deterministic chanciness to the trying-then-doing change it in any salient way? It doesn't seem like it would have anything to do with freedom, any more than your thumb involuntarily twitching in an uncaused non-deterministic way would have to do with freedom.

At that point, it seems a tautology to say that an event is both chancily uncaused and free.
posted by XMLicious at 3:40 AM on January 25, 2017


Or, it's probably better to say "an act": it seems a tautology to say that an act is both chancily uncaused and free.
posted by XMLicious at 3:46 AM on January 25, 2017


You can't take a thing out of the context that gives it meaning and then say you've proven it's meaningless by doing so. Of course a thing that depends on context to be meaningful looks meaningless when deliberately examined in isolation, outside its usual context.

The idea and term "meaninglessness" used as an absolute is meaningless, too, if the universe is meaningless absolutely. But if it isn't meaningless, then whatever our choices mean to us, that's what determines them, not clockwork or a coin flip.
posted by saulgoodman at 9:52 AM on January 25, 2017


If you start from the premise our own explanations and unconscious reasons play no role in determining a decision outcome because you start from a position that ignores or discounts the reality of the meanings of our conscious and unconscious thought processes and assumes them to be random, or at least arbitrary, mechanical or chemical operations, then yes, you can't see a possibility for free will to be compatible with determinism.

If you allow for the possibility of meaning to start with and examine that thought content, you could say the thinking drove the decision, which is what we all intuitively feel, while allowing that sometimes our thinking can go badly astray and even out of control. That's what I would argue.
posted by saulgoodman at 10:00 AM on January 25, 2017 [1 favorite]


I'm not seeking a way to make free will compatible with determinism, I'm simply asking what would free will be, what would the concept of free will signify, even in a non-deterministic universe?

Sure, you can define it as a meaningful outcome of conscious and unconscious thought processes, but either those conscious and unconscious thought processes are the sum of consequences of preceding external, initial conditions—because at the very least they're partly that, right?—or, what other factors might they be composed of?

I'm not taking any particular answer as a premise, but "chancey" genuinely-un-pre-determined randomness seems to be the factor we're discussing in asking how free will relates to whether the universe is deterministic or non-deterministic as relates to quantum phenomena.

Maybe there are third, fourth, nth options for the factors composing those meaningful outcomes of conscious and unconscious thought processes, but what are they, and what is the salient difference in including them versus meaningful outcomes of conscious and unconscious thought processes that are entirely the product of preceding external, initial conditions, as presumably would be the case in a deterministic universe?

(I'm having trouble parsing the statement that it's whatever our choices mean to us, that's what determines them, not clockwork or a coin flip, because it seems to be using a different definition for "determines" than would fit with the sense in which the universe is either deterministic or non-deterministic.)
posted by XMLicious at 3:38 PM on January 25, 2017


Thought processes are not selves. Selves can theoretically have emergent properties that you could never see while looking at the level of the parts, which gets back to my critique of that sort of analysis committing a category error.
posted by saulgoodman at 5:23 PM on January 25, 2017


Unless I missed it somewhere else in the thread, you're the one who brought up thought processes. It's possible to frame all of these questions as being about the characteristics of a unitary self (which we have to get into anyways because "has free will" is one of those characteristics) rather than anything about a proposed model of discrete parts of the self, so I don't think emphasizing the distinction resolves anything.
posted by XMLicious at 5:43 PM on January 25, 2017


I brought up thought processes as an example of constituent processes that go into making up the whole we call a "self"; my point is if you want to think and talk about how selves work as a whole, which is the level of abstraction/analysis at which it's been claimed free will exists, you can't decompose the self in your thinking if it's free will you're interested in thinking about or looking for, and not the composition of selves, which is a different kind of problem, operating at a different level of abstraction. All you need to prove the possibility of free will is proof that it's possible for a self to choose different options on different occasions from a range of possible choices given basically identical external conditions. That makes the self as a whole the decision agent in those cases and demonstrates the continent possibility, if not necessarily the absolute reality, of limited free will.
posted by saulgoodman at 9:27 AM on January 28, 2017


Heh. Missed the edit window. "Continent" = "contingent" but the dumb autocorrect in my phone is sneaky and seems to be optimized for selecting more commonly used words by default... Ach. Algorithmic dumbing down.
posted by saulgoodman at 10:03 AM on January 28, 2017


Again, though, I'm not trying to disprove the possibility of free will under the conditions you've laid out, I'm asking what free will is.

This loops us back to the beginning, which is an amusing thing to happen in a discussion of free will.

Put it another way: aren't the selves that make a different "choice" in each instance identical, identical even in their "emergent properties", as the product of identical initial conditions? Yet something must be different that factors into/causes the choice, if there is a different outcome. Whatever is different among the causes of the choice, whatever the delta is, it must itself be uncaused, i.e. it can't be traced back to the identical initial conditions, otherwise completely identical causes would produce identical outcomes.

So what is an uncaused thing that is not derived as a consequence from anything that has previously happened in the history of the universe, not from any of those initial conditions? It seems to me better described as a "coin flip" (or it could be more complex than a binary value, but random nonetheless) rather than "free will" or even "limited free will", if there is no reason for it and there can be no reason for it.

Even though this all satisfies the rules under which you've said free will is proved to be possible, I'm just inquiring into the nature of what exactly has been proved to be possible.
posted by XMLicious at 11:21 AM on January 28, 2017


Free will, as I define it, is just the ability to make one from a range of possible choices based solely on processes and states internal to the self.

Basically, for my definition, free will is no more or less than the idea that a self can choose one from a range of available options in the outside world based only on processes internal to the self, and that the same self when presented with identical options and external conditions can decide differently when presented the same range of options again. Free will is the ability of a self, viewed as a whole, to behave as the deciding agent for a particular outcome.
posted by saulgoodman at 5:46 PM on January 28, 2017


So, since you've repeatedly been presented with the notion and not denied it or qualified your proposition in response, I'm going to take it that the "internal processes" being random and uncaused in their selection among multiple outcomes still satisfies your definition of "free will", since as I've said it fulfills what you've presented as a proof of the possibility of free will.

So that's at least one type of free will we have here. It seems as though one could also say that the rules these internal processes operate by are such that, given the same internal starting state for the self, the same starting state evolved by the same set of rules can result in different outcomes. Maybe this capability for variation while operating under the same rules is one of the emergent properties of the self.

However, this seems functionally equivalent to the first type of free will, if there is no particular reason for the varying outcome and it's just something that happens sometimes.

My point is that I can't think of any other type of free will that would result in different outcomes when you re-wind and re-play time, other than something that's going to amount to random perturbations of those internal processes. If there were any reason that the choice was different in some particular instance, a way that the particular choice could be related back to some aspect of the unique-to-the-self-yet-identical internal initial state, or related back to some aspect of the unique-to-the-self-yet-identical and repeatable rules by which the internal processes operate, that connection must be tenuous enough that in other instances with the same initial internal state operating under the same rules it does not cause the same outcome of the choice.

And it seems to me that whatever breaks that tenuous connection differently in different instances must be random, because any other way I can think of explaining the differences would amount to different internal initial conditions or different rules under which they evolve. And then you're left with the question of why the internal initial conditions or rules they work under were different in that particular instance, which because we're working under identical external initial conditions as well, we're again left with the ultimate cause of a different choice being random and uncaused.
posted by XMLicious at 12:05 AM on January 29, 2017


I'm not trying to disprove the possibility of free will under the conditions you've laid out, I'm asking what free will is.

It's the grand unified theory of everything hiding in plain sight -- that grey area where things just don't add up, in philosophy, in psychology, in politics, in economics, mathematics, physics, music, EVERYTHING. The uncertainty at the heart of it all, which demands we juggle competing arguments and make some sort kind of decision as to what we should have for dinner.
posted by philip-random at 1:03 AM on January 29, 2017


But all that would also happen in a deterministic universe.

I'm okay personally with saying that the concept of "free will" is a phenomenon related to humans feeling uncertain, even in a deterministic universe; it seems fairly straightforward that free will would be something like that, in that case.

But saulgoodman and many other people state that free will has something to do with the possibility of non-pre-determined outcomes from the same identical initial conditions, and often explicitly define it in such a way that a mundane random variation in outcomes would qualify as a type of free will. I'm trying to get at whether free will can be anything more than that, even just in a purely conceptual sense, whether or not it can be proved as the genuine state of affairs in the universe we live in.
posted by XMLicious at 2:23 AM on January 29, 2017


Just pointing out those internal processes might be random doesn't prove it. I'm not going to assume thoughts are random and meaningless without proof, because that's an extraordinary claim.
posted by saulgoodman at 7:37 AM on January 29, 2017


And even if they were "random" as judged from some particular point of view--because judging something to be random versus meaningful is very much connected to point of view and context, too--yes, that would still satisfy my definition.
posted by saulgoodman at 7:39 AM on January 29, 2017


If you've been proceeding through this discussion with the understanding that whether or not something is random changes based on the point of view you look at it from, we may just be speaking completely different languages.

Looking at your profile it says that you're a software developer (as am I); I'm not talking about something like a pseudorandom number generator, which would be deterministic. (I don't know if you interpreted it that way.)

I think I used the word "meaningful" of my own accord once, way up above when I wasn't quoting you, but feel free to completely ignore whether or not anything is meaningful, because it's not relevant to and doesn't change anything I'm saying.

And I'm also not saying that thoughts are inherently random. "Randomness" as I'm using the term isn't something that would exist in a deterministic universe, and I haven't suggested that thought would not occur in a deterministic universe.

I started to take a shot at rewriting everything again using some other alternative to the term "random" and involving thoughts now that you're talking about thoughts again, but I guess I've run out of steam here. I feel like I haven't managed to communicate the question I'm asking, but thank you for engaging.
posted by XMLicious at 1:04 PM on January 29, 2017 [1 favorite]


there are mathematical facts that can be true for no reason at all (or that just elude understanding, clearly ;) and, perhaps, physical 'random' processes that may be entirely deterministic -- simultaneously determined? -- but unknowable (viz. given our brains, computational ability, etc.) whether that has anything to do with 'free will' or whatever i leave as an exercise for the reader!

also besides complex probability, i'd add exotic probabilities: "These are forms of probability theory that share many of the usual axioms of probability theory but in which the probabilities themselves lie in a set other than the non-negative reals eg. the complex numbers, the quaternions, or even the p-adics. The primary motivation is that classical mechanics plus complex probabilities looks a lot like quantum mechanics, and so if you believe in complex probabilities you no longer have to worry about things like wavefunction collapse. Unfortunately it's all a bit confusing if you're a frequentist." :P (cf. negative probability & octonionic probability!?)
posted by kliuless at 9:48 AM on February 11, 2017




« Older Eero Saarinen: The Architect Who Saw the Future   |   To the lighthouse! Newer »


This thread has been archived and is closed to new comments