Wednesday, January 13, 2016 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Free will of particles and people

It's so simple

This text is a simple attempt to rephrase the recent thoughts about free will (including the content of the comments) in a way that is more organized according to the philosophers' jargon.

Philosophers have debated "free will" for centuries. The basic question is whether free will exists but philosophers love to ask a different "first" question:

Is free will compatible with the determinism of the Universe?
Those who answer "Yes, compatible" are called "compatibilist". Those who answer "No, incompatible" are called incompatibilists. Now, many people have suggested that I was a "compatibilist", especially because I mention that the freedom may be defined as the absence of influence from other people or objects or, more precisely, the existence of some decisions that are not fully determined by those other people or objects.

The idea is that the existence of e.g. political liberty isn't some obscure metaphysical question but a very tangible, material question that decides about the human lives and the character of whole nations. I agree with that: freedom may be a bit vague concept but it's very real, too. I think that this agreement of mine is in no way equivalent to the compatibilists' defining proposition that "free will is compatible with determinism".

I am an incompatibilist because I really define free will to be the existence of data (free decisions) that are not determined by the earlier, external data (or agents). And I am confident that free will exists, even in principle. We will get to the evidence showing all these things.




The incompatibilists are divided to three basic groups:
  1. libertarians: free will exists, the Universe is not deterministic
  2. hard determinists: free will doesn't exist, the Universe is deterministic
  3. hard indeterminists: free will doesn't exist [thanks, Erwin and Bill], the Universe is not deterministic
In this sense – and, to some extent, even in some other (more political) senses – I am a libertarian. Free will exists because we feel it and the Universe, as most clearly implied by the intrinsic quantum randomness, is not deterministic. Yes, if one wants to talk about the strict free will which is free even in principle, it only exists because the universe is affected by fundamentally random data, the results of quantum measurements.




I also want to mention the issues about the particles' free will. But before we get there, let's discuss the reasons to trust the claims about the "free will of humans" above and their basic meaning – and show the mistakes of some basic counter-arguments.

We know that we have free will

First, the human behavior is often unpredictable, random. People often look stubborn. These are the "external manifestations" of free will of humans. But all the other people could very well be some machines or puppets that are controlled by some external puppet masters. The actual reason why I am sure about the existence of free will (and I mean my free will) is that I feel it. I know that many if not all of my decisions were done by me and not dictated by any external people or data or mechanisms. At least, I have eliminated all conceivable influences that could operate within the spacetime by mechanisms that at least remotely resemble those that I consider allowed by physics.

This basic comment relies on my subjective perception, something that I cannot prove. I cannot prove that I am aware of myself, I am conscious, and I know that a given decision was really mine. Only I know it for sure. People can say these things – but that's no proof. Tape recorders may say all these things as well – but they are not true. And liars may say lots of things that are not true. So the fact that someone says something is no evidence. It's simply hard to know what's going on in someone else's skull. Brain scans can detect something. But it seems totally conceivable to me that the "detailed ideas" will never be fully decoded. And maybe they will. But they haven't been.

All the evidence about the world – about anything – that I may really trust fundamentally boils down to my perceptions. Similarly, all the evidence that you should trust comes through your perception. You are observing the world but the information gets to your knowledge through your nerves and brain cells. In the Backreaction comment thread, Ppnl, Oren, and a few others were trying to convince Sabine Hossenfelder that aside from the "outside view", there also exists an "inside view" that involves feelings that simply can't be proven externally.

She denies the existence of her own internal feelings and consciousness of any sort that couldn't be communicated or empirically proven by external gadgets. To deny one's own consciousness sounds like something so incredibly stupid that I can't imagine that we are debating her. Maybe, she is a stupid robot without any feelings and consciousness. But we are not and for each of us, it is I or oneself whose perceptions we ultimately want to explain.

Sources of indeterminism in the world and in the brain

So the "libertarianism" is really a combination of two statements. I've mentioned why we take it for granted that we have free will. We just feel it. In our organization of the world and the separation of its data to "ourselves" and "the rest", with some meaningful sketch of the arrangement of "the rest", we simply know that our free decisions really come from "ourselves" and not "the rest".

Because I believe that all of us work in a similar way and we have common ancestors, I am confident that it's right for others to have a similar "inside perspective". I believe that if you're not insane, you are aware of your feelings. And at least if your brain is disconnected from various cables, you may be sure that your decisions weren't imposed upon you, they were created within you, probably your brain, and so on. That's why I think that no sane person should be interested in theories postulating or implying that "no free will exists" or "no consciousness exists" (these are two different things!).

The second part of "libertarianism" is the insight that the Universe is not deterministic. The Universe used to be believed to be deterministic in principle in the era of classical, deterministic physics. But this era has been over for more than 90 years. And even in classical physics, people had known about the importance of stochastic models with random terms at least from Einstein's explanations of the Brownian motion – and perhaps from the birth of statistical of physics in the 19th century, too.

This leads me to the sources of indeterminism – the impossibility to predict the evolution. In classical physics, things may be unpredictable or non-deterministic "in practice" because things are too complex, the evolution is chaotic, too accurate knowledge of initial conditions would be required, and so on (these propositions overlap: they are mostly equivalent). That's why the behavior of systems as complex as the human brain (if such objects were allowed by classical physics at all) would be "indeterministic in practice", even if it were "deterministic in principle".

Because I've also suggested that "free will" should be linked to the questions about the causal relationships – free will exists when some completely new data (decisions) "invented" or "born" within the system at the given moment have to be known, aside from the earlier and external facts, to predict the evolution of the system (or the Universe) – we must realize that the concept of "free will" will inevitably inherit some ambiguity from the fuzziness of the words "determined", "we", and a few others.

You know, the reduction of the term "free will" – which isn't a phrase you learn about in physics courses – to statements about the causal relationship between events is progress because physics talks about the causal relationship between events all the time. That's why it makes sense to decide about the existence of the "free will" within the framework of physics. However, even the more physics-friendly definition uses some words such as "inside us" and "determined" that may have several interpretations or stronger or weaker versions etc. Consequently, there will be various types of free will, too.

In the classical deterministic Universe, the motion of all particles – including those within our brains – would be determined by the laws of physics and the initial conditions. So in principle, whatever or whoever invented the initial conditions should be considered the "first cause" of all the events, including the apparent human decisions – if some events in that Universe could resemble those. The "people" wouldn't have the "free will in principle" in a deterministic Universe. That's why we are libertarians in principle.

On the other hand, they would have free will in practice. It's impossible for another real-world person to predict the behavior of complex enough objects such as the brain even if the underlying laws were deterministic. Such a brain could effectively behave much like a brain in our quantum mechanical world. There would be "seemingly random" aspects of its evolution all the time. In practice, it could have free will because the behavior would be in practice incalculable by others. But in principle, only the initial "clockmaster" would have free will in practice.

Now, does the quantum randomness play a role in the brain? You bet. Quantum randomness is everywhere and even if you applied decoherence and derived some effective classical equations of motion for the brain, they would have stochastic terms in them – which you could treat as classical stochastic terms, however (like the forces driving the Brownian motion). But decoherence shouldn't prevent you from seeing that the fundamental origin of these random terms is still quantum mechanical! Decoherence doesn't mean that the randomness goes away or quantum mechanics breaks down. It only means that the quantum randomness may be approximately (well enough) rewritten in terms of "classical probabilities" associated with preferred basis vectors of the Hilbert space (and the preferred basis means that you won't be able to observe any quantum interference in the decohered observables). But the probabilities are still numbers between 0% and 100%, like 96% and 4%.

Also, as I argued, tiny uncertainties of one neuron in the brain – and perhaps one elementary particle in it – are enough to influence the behavior of the whole human and, if he is an influential army officer, the extermination of whole nations, among other things. It's easy to see. The brain, like the CPU of a computer, is designed with the purpose of being able to amplify these small "decisions" done inside it so that they have a huge physical impact. Your nerves carefully "measure" some neurons in a part of the brain and these nerves are capable of commanding whole muscles. These muscles make the person speak and as he speaks, if he is a big boss in a military, he can send the whole navy to a different place, and so on. The amplification coefficient is basically unlimited. And the most consequential atoms are those in the brain or the CPU – which is why the brain and the CPU are command centers of the humans and computers, respectively.

The "just in practice" indeterminism may be removed in principle but the quantum mechanical randomness can't and it's the original source of lots of uncertainty in the brain, even if we often use the classical language to talk about it. There are lots of physical systems that show that the quantum randomness matters. Place a pencil on the tip in an unstable position. Both position and momentum will be uncertain up to \(10^{-17}\) or so in SI units; the uncertainty principle doesn't allow you more precise values. But the falling pencil (inverted harmonic oscillator) means that \(\Delta x\) as well as \(\Delta p\) is increasing exponentially so even with the minimum initial uncertainty, you are just "dozens of \(e\)-foldings" from the moment when the pencil falls completely.

What is "me"?

A person is said to have "free will" if some of the decisions – new data needed to predict the subsequent evolution – were random outcomes of measurements that took place within the body. Now, if a cell misbehaves, mutates, and starts to grow a tumor, was that the person's free decision?

It depends on how you define the "person". Because there is some ambiguity about which degrees of freedom in the human are considered "herself", there will be an unavoidable ambiguity about "her free will", too. The term "free will" can't be made more unequivocal than the other words it's made of. But because I wrote the clear definition, it may be (un)ambiguous to the same extent.

So if the tumor grows, the person will probably say that it wasn't her free decision. The tumor and the naughty cell do not belong to her body, she wants to believe, and they're not controlling her thinking which is the most important thing. Well, that's a wishful thinking of a part of the body, of course. If the tumor grows a lot, it may conquer the brain, too. Or at least damage it in other ways.

Different cells (and atoms) in the body and the brain coexist and largely cooperate. But it's not always the case. Some of the well-rounded (we say "full skinny") women may consider their accumulated fat cells to be invaders who don't belong there and these women wage a war against these cells of fat. Even when you think, you can make a mistake that may perhaps be blamed to some defective neuron, and so on. It's not really "you" who made the mistake. And those of us who have some eye disorder surely consider the eyes to be an imperfect external gadget we have to cooperate with.

This blame game may get complicated. The different cells (and atoms) resemble members of a nation or a society. The society often cooperates but the cooperation has its limits. The individual members of the society mostly play their own game – they are following their interests or performing their tasks, without directly planning consequences (especially not in some "big picture"; if someone plans the big picture, it's an important part of the brain). Is the terrorist attack by an Arab French citizen a free internal decision of France? It depends whether you count him as a part of "France". You may – or you may view him as a member of a special category that is largely waging a war against France (and the West).

Or take some temptations. You see a great bottle of an alcoholic beverage. It looks so great that you buy it and quickly drink it. You can't help yourself. Was it a case of your free will? Well, some instinctive part of your body connected to the nerves was able to convince your system to drink it. But you may believe that "you" are really some well-behaved brain cells in a part of the brain. Those nice cells didn't want to drink too much alcohol because it sucks. But these nice cells were bullied, defeated, and forced to behave by the naughty, pro-ethanol terrorist cells, you tell your physician, and you're partly right. Except that there's no good argument separating your "bad desires" of this kind from the "good ones".

Similar questions (I am sure you may invent as many examples as you want) apply whenever we talk about the "location" where the decision was invented. For this reason, "free will" may have various degrees depending on whether you include the instincts, typos caused by a neuron that just died, harmfully mutating cells, or unexpectedly decaying radioactive nuclei in your body to be "a part of you". It's a complicated business. But once you decide which influences you consider "determined" or "calculable" and what is "you", you may define your free will, too.

The demographics is predictable. Does it erase free will?

Kashyap and others have mentioned an example involving electrons but I will make the same one for humans because the situations really should be considered completely analogous. Take all "mature males" in Berlin – which happens to be 1 million people – and bring them a new sexual partner from Paris. It's a new EU campaign. Now, they may choose whether they want a same-sex partner or the opposite-sex partner. You complete the game and you will see that 4% i.e. 40,000 male Berliners chose a gay partner, the remaining 96% chose a woman.

Note that with some relatively small error, perhaps 200 or so, the number of Berliners who chose a gay relationship was determined by the known general laws of biology. Biology is messy but here we assume that it implies that 4% of men are gay as clearly as quantum mechanics implies probabilistic predictions for the electron's spin in a given experiment.

Because the percentage of gays is basically known in advance and the Berliners were assigned "randomly", does it mean that they don't have free will? Sabine Hossenfelder basically says "Yes". But it is a complete and silly misunderstanding of free will. The overall percentage is just one number. But the assignment of the sex partners to the male Berliners involves 1 million additional bits of data! Those are not irrelevant. They are the bulk of the information in the problem.

What Hossenfelder and others imagine is that she only needs to reproduce the "overall percentages" so the free decision of the Berliners may be replaced by a centrally assigned type of sex partners. Angela Merkel's cell phone runs its good enough pseudorandom generator 1 million times. It produces numbers between 0 and 1. The distribution is uniform. If you get less than 0.04, you will get a gay partner. You will be obliged to say "I demand a gay partner from my beloved German government and thank my French comrades for him." Above 0.04, the same procedure follows with the word "straight" and "her".

This is great: with this smartphone solution, the Berliners don't have to think anymore and the overall numbers agree. 40,000 of Berliners will be made to sleep with men. So Hossenfelder explicitly says things like:
If it is random, there is no agency to it, consequently there is no "will".
Just because the individual decision is random, in the sense of being a contribution to a statistical ensemble with predictable macro-properties, there is no agency to it? Wow. Tell it to the (4% of 96%) 38,400 straight Berliners who will be forced to sleep with other men and (96% of 4%) 38,400 gay Berliners (almost all gays) who will be forced to sleep with women!

The point is that the overall bureaucratic data may look the same if you replace the individual random generators by some external ones (or perhaps a pseudorandom generator). But the people whose will was overwritten by external influences will see the difference between a free decision and a non-free decision very well, especially if they belong among the 76,800 Berlins who got a "wrong type" of a partner.

Now, the troubles resulting from the wrong assignment may be visible. But they may also be made very or completely invisible. No one should look into your bedroom. And you may always pretend that you think and want something else than you do. And the assignment of the partners could have occurred at the first moment when the males, as boys, were deciding about their orientation. But the key point is that you know damn well whether your decision was free or imposed upon you, whether it agreed with what you really wanted. As Bruce Rout mentioned,
1) An agent in possession of free will is able to perform an action that was possible to predict by nobody but the agent itself.
This is exactly right. Free will exists when (and because) the subsequent evolution of your body may only be predicted by yourself because you know what you're gonna decide in a split second. And your body may influence many other things in the world – which is why your free will is so important.

But your free will (and what it actually wants to do) is something that only you may really feel certain about.

Free will of particles

Conway and Kochen have proven that if we assume the right spectrum of a certain spin-one observable; the existence of the entangled states of two spin-one particles (which produce the usual correlated results of measurements); and the impossibility to send real superluminal influence between the particle and the other experimenter (some combination of relativistic locality and free will of the humans), then it follows that the elementary particles have a free will, too.

This is a "poetic" way of saying that the results of their measurements can't be predicted from the knowledge of any past or external data.

Now, if you think that the "free will of particles" sounds unusual if not comical, be sure that it sounds unusual or comical to me, Conway, and Kochen, too. It's simply not the kind of language we normally use – neither in everyday life nor in physics. But this "bizarre" character of the phrase "particles' free will" doesn't prove that particles don't have free will.

(If you "define" free will as something that can only be talked about in the context of humans, or if you "define" it to be non-existent for non-humans, be my guest. But it's a stupid, contrived, and potentially self-contradictory definition – just like if you try to "define" that only humans may be "short". Well, it's simply not true: ropes and months may be short, too. You can't "define" falsehoods, not even falsehoods that sloppy people who haven't thought carefully whether the particles have free will have been telling you many times.)

We may define free will in the "search for the first causes and check if all of them are external" way – and this way is arguably needed to understand the human free will in a way that is compatible with the physical description of humans. (One may also understand why it's a good idea to link this kind of a free will with the responsibility for crimes and all other acts.) But once you have such a definition, it turns out that you may apply it to unusual systems including the elementary particles, too.

And when you do, and if you carefully analyze what's happening to the particles during their "life" and the subsequent "measurement", you will see that the right answer is that the particles have free will, too. It's really a theorem they have proven.

Let me finally say that none of the things above is taught in any physics classes. There are no equations that are too helpful for the understanding of the broad picture. People may prefer to talk about these things differently. And that's why physicists generally avoid any discussions about similar, deeply metaphysical issues.

10+ stupidities symbolized by Ms Hossenfelder

On the other hand, despite the non-existent exposure of physics students to this stuff, I am confident that it is objectively stupid to deny her own consciousness or her own free will (which are different things but some people deny both).

It is stupid to say that one's free will can't be distinguished from the will imposed on you by others because all of us really directly "feel" the difference. It's also stupid to say that particles don't have free will without a tiny glimpse of a rational evidence. (The truth is demonstrably the opposite, particles do have free will.) Also, it's stupid to say that people aren't free to push certain buttons when 2 external particles outside their bodies happen to be a in a certain state. (This "superdeterminism" would be a brutal violation locality as well as the ultimate conspiracy theory.)

It's just plain idiocy for someone to say that one's decision isn't free just because it has a random aspect – it's free exactly because it has it. It's totally hypocritical to call for equations and use the authority of physics – while rejecting every single important principle that physics has discovered and believing that the Universe may be described by some non-relativistic, fundamentally non-quantum, superdeterministic theory (and certain people promote these adjectives explicitly).

It's unbelievable for one individual to accumulate all these stupidities at the same moment. But Ms Hossenfelder has managed to get this 0.00% score in correctness about these conceptual issues, anyway.

Add to del.icio.us Digg this Add to reddit

snail feedback (0) :