Is determinism truly at play in this scenario?

[AlephZero]

To give an illustration from computer science, the halting problem: Some computer programs get into an infinite loop and never halt. That's completely predetermined by the program. But you can show (this was shown by Turing) that there is no way to predict which computer programs halt and which ones don't.

That is slightly wrong, but the "slightly" seems important to this discussion.

Of course you can (almost always) predict whether any particular computer program will halt or not, by applying the ordinary rules of logic.

What Turing showed was this: you can't produce just one algorithm (which you could implement as a computer program) that will predict whether all possible computer programs will halt or not.

The proof is very simple: suppose such a "magic debugger" exists. Now write a program that says "If the magic debugger says this program will loop for ever, then halt, otherwise loop for ever".

But how this relates to determinism is not so simple IMO. The proof says nothing about you can prove by some method or other whether all possible programs halt or loop. All it says is that one particular string of symbols in a programming language (i.e. the one in quotes in the previous paragraph) is not actually a computer program, it is only a meaningless string of symbols that looks similar to a program.

 

[craigi]

But we aren't talking about determinism. We are talking about superdeterminism.

Superdeterminism is a theory in which local hidden variables explain correlations between random measurement choices and results even though the choices and results occur at spacelike separation. The "choice" here has nothing to do with human free will. In these experiments, we delegate the "random choice" to an arbitrarily complex device whose detailed workings and initial conditions are unknown to us, eg. whether the number of raindrops that falls in a certain time is even or odd.

Determinism could certainly be true, and Zeilinger's objections just seem to apply to determinism. Superdeterminism could also be true. But can we really construct a predictive theory ("small" number of parameters) that is deterministic, and makes distant correlations possible depending on whether the number of raindrops or some other absurd parameters like the number of atoms in a ball of earwax at distant locations is even or odd?

In one case, we can know that superdeterminsim is viable.

Whatever the true nature of the universe, in principle, we can imagine taking a record of the result of every event. If we can do this, then we know this record could be referenced by a superdeterministic lookup mechanism.

Obviously, this isn't a useful model of the universe because it lacks predictive power. We like to be able to, measure, or at least prepare, the initial state of an isolated system, then use nothing more than rules to predict its future state.

 

[atyy]

In one case, we can know that superdeterminsim is viable.

Whatever the true nature of the universe, in principle, we can imagine taking a record of the result of every event. If we can do this, then we know this record could be referenced by a superdeterministic lookup mechanism.

Obviously, this isn't a useful model of the universe because it lacks predictive power. We like to be able to, measure, or at least prepare, the initial state of an isolated system, then use nothing more than rules to predict its future state.

Yes, I think that's the problem with superdeterminism - it could be true, but for us to know it we'd have to do so many experiments to find out the parameters, that we'd run out of predictive power.

Even with more reasonable theories like dBB, we already seem to have a huge number of possibilities (http://arxiv.org/abs/0706.2522) - none of which are distinguished - unless QM fails. If QM fails we do have a lead. But dBB and its variants are interesting for the moment because they conceptually solve the measurement problem, not because we actually want to use it (ok, it's useful for some calculations), but basically we can believe in dBB for ontology and use QM for predictions. Here's an example showing that a dBB-like theory must be much more complex than its QM counterpart http://arxiv.org/abs/0711.4770.

BUT: I need to understand what Hossenfelder is saying about some class of SDHV theories being testable, to see if there are exceptions.

 

[Pythagorean]

But I don't see why anyone would be a determinist about inanimate nature, but not our own behavior. Superdeterminism appear to follow automatically. The universe is the same for each and every moving subject.

Determinism isn't about animate/inanimate. It's about the nature of the process itself. Some processes are just stochastic, some are just deterministic. And we use these words, really, to describe our modelling approach more than anything. A really complicated deterministic system can also be modeled stochastically, lending to the confusion.

I don't really understand what superdeterminism is so I won't comment on that.

 

[craigi]

I need to understand what Hossenfelder is saying about some class of SDHV theories being testable, to see if there are exceptions.

She's just taking a metric of a correlation time for a particular particle, and proposing testing an upper bound on it.

Short correlation time would map to high complexity, in that the particle would be changing state, for interaction, at a high rate.

 

[stevendaryl]

So would you agree that superdeterminism is not "determinism taken to its extreme"?

As I understand the role of "superdeterminism" as a loophole to Bell's inequality is that it has to do with independent choices.

In a hidden-variables model of the EPR experiment, there are three choices that must be made, by three different "agents":

1. Alice chooses her detector setting.
2. Bob chooses his detector setting.
3. Nature chooses a "hidden variable" shared by the twin pair of particles.

The issue is whether these choices are made independently, or not. If Alice's and Bob's choices were somehow predetermined, then the hidden variable could be chosen taking their choices into account, and you could reproduce the predictions of QM.

The reason this goes beyond ordinary determinism is that you could arrange for the three choices to be independent, even if determinism holds. That is, you can assume that the initial conditions for different parts of the universe could be chosen independently.

 

[atyy]

Sabine Hossenfelder certainly doesn't think so.
http://arxiv.org/abs/1105.4326

Here's a paper that seems related.

http://arxiv.org/abs/1205.0878
Are quantum correlations genuinely quantum?
Antonio Di Lorenzo
Int. J. Mod. Phys. B, 27, 1345016 (2013)

 

[OCR]

It was Wheeler again, who proposed the reason that elementary particles of the same type are indistinguishable, is that they are actually the same particle wound round in space and time. Antiparticles being particles traveling backwards in time. Feynman raised the objection that we don't see as many particles as antiparticles and Wheeler offered some hand-wavey argument.

Feynman actually said... "But, Professor, there aren't as many positrons as electrons."

Wheeler responded with... "Well, maybe they are hidden in the protons or something".

http://en.wikipedia.org/wiki/One-electron_universe


Carry on...

OCR

 

[bhobba]

The universe is governed by the laws of quantum mechanics. The equations of QM are deterministic, the observations are indeterministic.

That is a matter of interpretation.

We have interpretations such as Nelsons Stochastic's and Partial State Diffusion where QM is not deterministic and interpretations where everything is deterministic eg MW and DBB.

As far as we can tell today there is no physical theory that, if its deterministic can't be derived from some probabilistic basis (eg Classical Mechanics from QM) and conversely (eg QM from a deterministic theory).

This makes such discussions, IMHO useless philosophical waffle, until some way can be found to decide the matter experimentally - which IMHO is highly doubtful - but one never knows what future progress may bring. Just my view - chat away about it as much as you like - its valid intellectual enquiry - just not to my taste.

Thanks
Bill

 

[bhobba]

Determinism isn't about animate/inanimate. It's about the nature of the process itself. Some processes are just stochastic, some are just deterministic. And we use these words, really, to describe our modelling approach more than anything. A really complicated deterministic system can also be modeled stochastically, lending to the confusion.

Exactly - which is why such discussions to me don't really seem to go anywhere, and even as a matter of principle I suspect they cant. Still one never knows.

Thanks
Bill

 

[craigi]

Exactly - which is why such discussions to me don't really seem to go anywhere, and even as a matter of principle I suspect they cant. Still one never knows.

Thanks
Bill

For the largest part, it's all the same physics, just different ways of looking at it. As you suggest, understanding them has its own reward. Amongst many quantum physicists, there still seems to be a feeling that they haven't found a satisfactory view on the subject. Surveys show a suprising disparity between the interpretations that they hold to and it is still an active area of research.

I think many people take in interest in physics because they hope to achieve a deep understanding of nature. I don't find that stopping at the QM formalism or what are referred to as a minimalist interpretations fullfils that. For me, it left me with many more questions that I started with.

QM has proven to provide very reliable predictive power, but there have been minor modifications over the years. Different interpretations can yield, or at least hint at, where new testable predictions might lie, even if they're beyond our means at the moment.

I think the biggest relevant issue is that we've made very little progress in creating a theory of quantum gravity and it seems to me at least, that intrepretational issues present a serious problem for that.

 

[atyy]

Determinism isn't about animate/inanimate. It's about the nature of the process itself. Some processes are just stochastic, some are just deterministic. And we use these words, really, to describe our modelling approach more than anything. A really complicated deterministic system can also be modeled stochastically, lending to the confusion.

I don't really understand what superdeterminism is so I won't comment on that.

That is a matter of interpretation.

We have interpretations such as Nelsons Stochastic's and Partial State Diffusion where QM is not deterministic and interpretations where everything is deterministic eg MW and DBB.

As far as we can tell today there is no physical theory that, if its deterministic can't be derived from some probabilistic basis (eg Classical Mechanics from QM) and conversely (eg QM from a deterministic theory).

This makes such discussions, IMHO useless philosophical waffle, until some way can be found to decide the matter experimentally - which IMHO is highly doubtful - but one never knows what future progress may bring. Just my view - chat away about it as much as you like - its valid intellectual enquiry - just not to my taste.

Thanks
Bill

The point is not determinism/randomness - the point is locality. We are interested in theories with hidden variables because these solve the measurement problem in a straightforward way. Then the question is whether we can have a local hidden variables theory that reproduces the predictions of quantum mechanics. And if one does have such a local hidden variables theory can it also be predictive?

 

[bhobba]

The point is not determinism/randomness - the point is locality. We are interested in theories with hidden variables because these solve the measurement problem in a straightforward way. Then the question is whether we can have a local hidden variables theory that reproduces the predictions of quantum mechanics. And if one does have such a local hidden variables theory can it also be predictive?

All this is valid intellectual enquiry.

For me though until experiment can decide its all a matter of opinion and taste - not science.

Thanks
Bill

 

[bhobba]

I think the biggest relevant issue is that we've made very little progress in creating a theory of quantum gravity and it seems to me at least, that intrepretational issues present a serious problem for that.

Very hard to know what future research will bring. But our best current theory for accomplishing it, String Theory, makes use of bog standard QM.

However I do believe eventually it will have something important to say about the foundations of QM - just a gut feeling though.

Thanks
Bill

 

[Fredrik]

Taking to it's extreme, determinism would seem to require irrationality. Suppose that I live in a deterministic
universe, all events being determined, experiments, brain states, cosmology and so forth. Suppose further that I possesses psychic abilities(perfectly possible in a relativistic-quantum universe) and I precognize an accident involving myself in 5 seconds.

In such a deterministic universe, I would prohibited from intervention, regardless of my true beliefs, and rational justifications for those beliefs. For the simple fact that the precognized event must come about.

I tried to come up with a thought experiment where I can see the problem with deterministic theories that you're talking about. But I wanted to remove the psychic powers, which frankly sound like nonsense to me, so I decided to replace them with computation of future events from measured results. I don't see any problems.

Consider the following thought experiment: Alice and Bob live in a universe that is perfectly described by a deterministic theory. Alice is in her spaceship deep in intergalactic space. Bob (on the outside) measures the state of Alice's ship and everything inside of it. He then uses the theory to compute that Alice's ship is going to explode. He sends a message to inform Alice, and she then pushes the emergency shutdown button, so that the ship doesn't explode. This scenario doesn't contradict determinism, because Bob's calculation only told him what was going to happen if Alice never receives the message.

So let's try to change something in this scenario in order to perhaps reach some kind of absurdity.

Let's say that Bob, immediately after sending the message, calculates how it will change the state of Alice's ship, and then calculates what will happen to the ship. This time he finds that Alice receives the message, pushes the shutdown button, and doesn't explode. Or maybe he finds that Alice's receiver malfunctions, so Alice never gets the message and then explodes. There's nothing absurd about these scenarios. It would be absurd if he finds that Alice receives the message and chooses to let the ship explode, but why would we assume that there are such solutions to the equations of motion?

Now let's remove Bob from the picture altogether, and say that Alice measures the state of her ship herself, and uses her own ship's computers to perform the calculation. This can only lead to an absurdity if the theory says that a physical system can determine its own state. I'm not sure how this would work. The ship's computer is much less complicated than the ship itself, and yet it has to store all the information about the ship. That sounds impossible in principle. But just to make things as messed up as possible, let's assume that we're dealing with such a theory, and that the computer has stored all the information that it needs to begin the calculation. There's only going to be a problem if the ship's computer is able to finish the calculation of what's going to happen 5 minutes from now in less than 5 minutes. This may also be impossible in principle. But let's just say that it's not, and that the computer spits out a description of what's going to happen before it happens. Again, there's no reason to think that the equations of motion have solutions that describe irrational behavior. It's still possible that all solutions are variations of these two: 1. Alice sees the result and pushes the button. No explosion. 2. Alice doesn't see the result and doesn't push the button. Kaboom.

I don't think the stuff in the last paragraph (where Alice is doing the calculation herself) is even worth talking about, because to do so, we have to speculate that physical systems can not only find out what their states are, but also store all the information and use it to compute what's going to happen faster than it's actually happening.

 

[eloheim]

Consider the following thought experiment: Alice and Bob live in a universe that is perfectly described by a deterministic theory. Alice is in her spaceship deep in intergalactic space. Bob (on the outside) measures the state of Alice's ship and everything inside of it. He then uses the theory to compute that Alice's ship is going to explode. He sends a message to inform Alice, and she then pushes the emergency shutdown button, so that the ship doesn't explode. This scenario doesn't contradict determinism, because Bob's calculation only told him what was going to happen if Alice never receives the message.

So howcome Bob didn't just calculate Alice's brain states as well, and use simulations to ensure that his message would be interpreted properly and begin the correct cascade of events leading from Alice's brain correctly deciphering the note to initiating the necessary motor movements for push the off button, to the ship responding properly and the explosion being averted?

I may have more to say about your post later because I think there's some interesting fodder in there...something almost makes me think of the halting problem incarnate...

 

[stevendaryl]

So howcome Bob didn't just calculate Alice's brain states as well, and use simulations to ensure that his message would be interpreted properly and begin the correct cascade of events leading from Alice's brain correctly deciphering the note to initiating the necessary motor movements for push the off button, to the ship responding properly and the explosion being averted?

I may have more to say about your post later because I think there's some interesting fodder in there...something almost makes me think of the halting problem incarnate...

I think the halting problem is exactly relevant. What the halting problem proves is that even when everything is perfectly deterministic, it may be impossible to make precise predictions in certain circumstances. You can actually force such a situation to occur by attempting to create a paradox.

 

[eloheim]

Different course (from my previous post) for a second. I hope I'm not missing something major here as far as how 'superdeterminism' goes, ...but can someone explain to me how exactly systems with no shared history or interaction could 'know' how to conspire together to produce the kind of correlations we see in entanglement games, etc.? Mustn't there be some physical mechanism or explanation?

I read the 'bell loophole' argument as being for a way to account for bell stats--while keeping the EPR premises (partially?) intakt--or no? Or does 'superdeterminism' imply each particle has access to any/all information about other happenings in the universe, regardless of distance or time? Because I don't see how else or why causality and appearances would be kept up so carefully in a universe as complex as ours.

 

[Fredrik]

So howcome Bob didn't just calculate Alice's brain states as well, and use simulations to ensure that his message would be interpreted properly and begin the correct cascade of events leading from Alice's brain correctly deciphering the note to initiating the necessary motor movements for push the off button, to the ship responding properly and the explosion being averted?

I was talking about measuring "the state of Alice's ship and everything inside of it", so this includes her brain state. Bob then calculates what the ship and everything inside it will do for the next five minutes or so. I'm assuming that he knows exactly how every possible message he can send will change the state of the ship and everything inside it, so that he can use the changed state as the starting point of a new calculation. Since we're talking about a deterministic theory, there's exactly one solution to the equations of motion (one thing that Alice and the ship can do) for each message that Bob can send. But there's no reason to think that any of them describes a future where the message is received and understood, but still ignored.

If there is such a solution, then yes, the universe that Alice and Bob live in is such that Bob can make Alice behave in a self-destructive way by doing something that we would think is completely unrelated to what she ends up doing. It could be something as silly as including a specific code in the message that Alice doesn't even see is there. It would be appropriate to call such a code "a spell" and the result "magic". As long as there's no evidence of anything like that in our universe, I'm not going to spend too much time worrying about its possible existence.

The point of my argument is that weird behavior ("magic") isn't a result of determinism. If there is such a thing in a deterministic universe, it's a consequence of the existence of "magical" solutions to the equations of motion, not a consequence of the fact that the theory is deterministic.

 

[craigi]

I was talking about measuring "the state of Alice's ship and everything inside of it", so this includes her brain state. Bob then calculates what the ship and everything inside it will do for the next five minutes or so. I'm assuming that he knows exactly how every possible message he can send will change the state of the ship and everything inside it, so that he can use the changed state as the starting point of a new calculation. Since we're talking about a deterministic theory, there's exactly one solution to the equations of motion (one thing that Alice and the ship can do) for each message that Bob can send. But there's no reason to think that any of them describes a future where the message is received and understood, but still ignored.

If there is such a solution, then yes, the universe that Alice and Bob live in is such that Bob can make Alice behave in a self-destructive way by doing something that we would think is completely unrelated to what she ends up doing. It could be something as silly as including a specific code in the message that Alice doesn't even see is there. It would be appropriate to call such a code "a spell" and the result "magic". As long as there's no evidence of anything like that in our universe, I'm not going to spend too much time worrying about its possible existence.

The point of my argument is that weird behavior ("magic") isn't a result of determinism. If there is such a thing in a deterministic universe, it's a consequence of the existence of "magical" solutions to the equations of motion, not a consequence of the fact that the theory is deterministic.

A deterministic system behaves deterministically in isolation. If a system can be modifed by an external indeterministic system then it is not intrinsically determinisitic, though it may behave deterministically until modified.

A deterministic system can be separated into deterministic subsystems. If the entire universe were to be deterministic, then in your example, neither Alice or Bob could change their predestined decisions. It would appear to both Alice and Bob that they were making decsions based upon the information they had, but those decisions and that information would be predetermined.