Is causality challenged by quantum theory's reliance on indeterminacy?

[Borean]

Hello all.

I am merely an inquirer into physics; not even an amateur.

The question that has puzzled me with regard to quantum theory/physics/mechanics, is the status of causality.

What is the status of causality in quantum theory?

I would be grateful if anyone could explain this to me.

Borean

 

[Bob_for_short]

Everything is OK with it: the wave function equation is causal, its solution is determined with the initial conditions, just like in classical mechanics.

Another thing is a probabilistic outcome of the measurement process. The measurement "points" are spread out (remember an interference picture). But it is an intrinsic property of quantum world. The average (inclusive) things are well predicted with the wave function.

 

[Borean]

My question is more specifically if quantum theory is acausal.

I read in another discussion that John Bells experiments and Bells theorem means that we are forced to abandon causality. Here is the discussion: http://www.physicspost.com/physicsforums/topic.asp-ARCHIVE=&TOPIC_ID=1602.htm

 

[alexepascual]

Quantum theory has two components. There is the wave function (or state vector) which is assumed to evolve in a causal way before it collapses, and then there is the collapse process which is not causal.
There is a standard mathematical machinery that describes these processes and which has been proven to work in all attempted experiments. This mathematical formulation gives good predictions but the results are of a probabilistic nature.
While the mathematical formulation is usually not challenged due to its proven success, its interpretation remains ope to debate.
You can use the math to predict (to the extent allowed by the theory) the results of the experiments. But when you start asking: What does this really mean? you'll get mainly two different answers:
(1) You should not ask this question. Just keep using the math (because it works).
(2) It is valid and potentially useful to try to gain a better understanding of how the different elements in the mathematical apparatus of quantum mechanics correspond (or not) to elements of reality.

 

[Bob_for_short]

I read in another discussion that John Bells experiments and Bells theorem means that we are forced to abandon causality. Here is the discussion: http://www.physicspost.com/physicsforums/topic.asp-ARCHIVE=&TOPIC_ID=1602.htm

It only means that quantum world is different from classical one.

The main thing - the wave function evolution - is causal.

 

[Borean]

Quantum theory has two components. There is the wave function (or state vector) which is assumed to evolve in a causal way before it collapses, and then there is the collapse process which is not causal.
There is a standard mathematical machinery that describes these processes and which has been proven to work in all attempted experiments. This mathematical formulation gives good predictions but the results are of a probabilistic nature.
While the mathematical formulation is usually not challenged due to its proven success, its interpretation remains ope to debate.
You can use the math to predict (to the extent allowed by the theory) the results of the experiments. But when you start asking: What does this really mean? you'll get mainly two different answers:
(1) You should not ask this question. Just keep using the math (because it works).
(2) It is valid and potentially useful to try to gain a better understanding of how the different elements in the mathematical apparatus of quantum mechanics correspond (or not) to elements of reality.

So that means that the math, in itself, is neutral to whether or not there is causality or not behind our observations?

Or does the math specifically force us to abandon causality?

Since we can still predict this supposedly "acausal" wave collapse process probabilistically, wouldn't it be incorrect to say that it is "acausal", and more correct to say that it is not fully determinable?

Causality really just means one thing depending on another, even if we don't know what depends on what, and even if we can't explain every causative event causally, and even if we don't have the observational abillity to give a causative account of everything in the universe.

That is, causality is not = determinism.

 

[Bob_for_short]

There is no wave function collapse while measurement as there is no probability collapse. The notion of probability belongs (covers) all (big amount of) measurements.

 

[Borean]

There is no wave function collapse while measurement as there is no probability collapse. The notion of probability belongs (covers) all (big amount of) measurements.

Yes.

But is the "wave function collapse" (or anything else in quantum theory) explicitly acausal or is it simply undeterminable what it's "cause" is?

 

[Bob_for_short]

But is the "wave function collapse" (or anything else in quantum theory) explicitly acausal or is it simply undeterminable what it's "cause" is?

The cause is an interaction with a measurement device. If we speak of photon absorption by a photographic plate, it cannot be predicted because the plate consists of many atoms whose sizes are much smaller than the photon wave-length. Who knows which one is the most ready to absorb a photon? So such events are not predictable but only their sum.

 

[Borean]

The cause is an interaction with a measurement device. If we speak of photon absorption by a photographic plate, it cannot be predicted because the plate consists of many atoms whose sizes are much smaller than the photon wave-length. Who knows which one is the most ready to absorb a photon? So such events are not predictable but only their sum.

Ah, I see.

So there are some limitations as to our observatory power.

But my question still remains; are there any explicitly acausal things in the universe, according to quantum theory? Can things happen without causes?

And remember, indeterminacy does not mean acausality. If things happen without us knowing the cause, that does not mean they have none. What I am asking is rather if quantum theory positively affirms acausality as such.

 

[Bob_for_short]

No, as far as I know, there is no such an affirmation. There is just huge difference in classical mechanical and quantum mechanical causalities, but there is no acausality in QM.

I would say that the classical mechanical determinism emerges from the inclusive QM picture when many-many events are added together and averaged. The average value is unique.

 

[Borean]

No, as far as I know, there is no such an affirmation. There is just huge difference in classical mechanical and quantum mechanical causalities, but there is no acausality in QM.

I would say that the classical mechanical determinism emerges from the inclusive QM picture when many-many events are added together and averaged. The average value is unique.

Interesting. What about something like atom/nuclear decay? Isn't that acausal?

The discussion I linked earlier says that it is.

Let me quote:

Question: Is it true that quantum mechanics allows/requires certain processes to occur without cause?

Answer: Yes, in a probablistic fashion. For example, if there are two possible decay paths for an unstable particle--it can decay into x+ and y- or into s0 and t0, then there is nothing--nothing--that causes it to take one path or the other. The path taken by the decay is acausal. Likewise, we can say that the mean life of a free neutron is only 15 minutes, but one particular neutron may decay in 14 minutes and another in 16 minutes, and there is no causality connected with the difference in specific decay times. An electron may have a 12% chance of tunnelling across a gap (like a tunnel diode) in .01 seconds, but nothing causes any particular electron to tunnel or fail to tunnel.

It's important to note that physicists are as certain as they can be that there are no hidden variables which control these acausalities, "behind the scenes" as it were. John Bell proposed an experiment which would, according to his analysis, reveal whether or not any such set of variables could be influencing quantum uncertainty. When performed, the experiments indicate that there is no set of hidden variables that are linked to quantum events: the quantum incorporates an element of genuine, irreducible uncertainty, or acausality in the sense described above.

It's very interesting that current best theory says that the entire universe once occupied a region comparable to the Planck size, and thus the entire universe would have been subject to quantum uncertainty.

--Don

But I suspect what he really means is that it is indeterminable, to us.

Can anyone clarify this?

 

[Bob_for_short]

Interesting. What about something like atom/nuclear decay? Isn't that acausal?...But I suspect what he really means is that it is indeterminable, to us.
Can anyone clarify this?

It is the same song. One cannot predict any single quantum event, it is a nature feature. But you can:

1) predict whether it will happen or not. The radioactive decay law says it will happen for sure.

2) predict the average number of decayed or remaining nuclei N(t) in a sample of many-many atoms.

Any single event, taken separately (observed only one time), does not even say what has happened. You need statistics to create notions, language of phenomenon description.

The radioactive decay has a clear cause - the potential barrier transparency (for simplicity).

 

[Borean]

It is the same song. One cannot predict any single quantum event, it is a nature feature. But you can:

1) predict whether it will happen or not. The radioactive decay law says it will happen for sure.

2) predict the average number of decayed or remaining nuclei N(t) in a sample of many-many atoms.

Any single event, taken separately (observed only one time), does not even say what has happened. You need statistics to create notions, language of phenomenon description.

The radioactive decay has a clear cause - the potential barrier transparency (for simplicity).

So does that mean that single quantum events don't have any cause -are acausal- or simply that they are indeterminable/unpredictable in isolation but still have a cause?

 

[Bob_for_short]

So does that mean that single quantum events don't have any cause -are acausal- or simply that they are indeterminable/unpredictable in isolation?

They are are indeterminable/unpredictable but have a certain cause to happen.

Look at the determinism as at an average (unique) picture obtained from many spread, occasional points.

 

[gel]

QM is acausal, as explained by the following experiments.

Shine light through a polarizing filter. Some fraction will pass through it, which will then be polarized.
Now place a second polarizing filter after the first, and aligned with it. Most of the light which gets through the first filter will also pass through the second (100% in theory, but polarizers are not perfect).
Rotate the second polarizer to that it is aligned at 90 deg to the first. Then, none of the light will get through it.
Now align the polarization of the second filter at 45 deg to the first and about 1/2 the light will get through.
However, you can reduce the intensity of light so far that there is only a single photon passing through the apparatus at anyone time. Given that the photon passes through the first filter, it is not possible for "half the photon" to pass through the second. Instead, QM says that the photon either passes through the second filter or it doesn't. The outcome is random, with each possibility happening with a probability of 50%.

When it is said that the wavefunction is deterministic, this means that we can calculate the 50% probability precisely (the probabilities themselves are deterministic), but the outcome of the experiment is still random.

According to the standard interpretation of QM, the outcome is fundamentally random and it is theoretically impossible to predict the outcome.
Of course, you can always suggest that maybe there is some more fundamental theory than QM with additional underlying variables, the knowledge of which would allow you to accurately predict the outcome of such experiments. However, there are good reasons for believing that this is not the case (e.g., Bell's theorem), although it doesn't convince absolutely everyone.

 

[alexepascual]

QM is causal only to the extent that the wave function can accurately predict probabilities. But that is not what we normally understand as strict causality. Let's say we have a wave packet that represents a particle. We can predict how the shape and the position of the wave packet is going to evolve. But when we try to find the location of the particle, we can't say with certainty were we are going to find it. In this respect QM is acausal.
You can send a photon through a half-silvered mirror and it'll have 50% chance of being reflected and 50% chance of being transmitted. You can put detectors that will find out if it was reflected or transmitted. There is no element in the theory that allows you to know beforehand which outcome you are going to get. This is acausal.
There have been discusions about the possibility of some hidden variable determining the outcome of apparently random events. I think it was determined that if there were hidden variables they could not be local.
Nobody has ever been able to exploit the concept of a hidden variable to predict the result of an otherwise random event.
The idea of random (probabilistic) events does not need to be that unacceptable. When you travel through space, you cas find objects scattered around without any "causal" relationship to the adjacent objects. By looking at a region of space you can't predict what you'll find in adjacent regions. Why can't it be the same with time?.
If you believe that there is "one world" (that is one past, one present and one future) you could picture this world as represented in a strip of film. You will find some connection between adjacent frames in the film. Usually you'll see that objects "move" smoothly, etc. But it doesn't necessarily have to be that way. You could imagine a film where a particle is at a totally random location in each frame. In this case you would be assuming that the particle does have a particular location which is unpredictable. But I think there are experiments that show that even assuming that the particle moves in an acausal way is not enough. You have to assume that it does not have a particular location until you measure it.
Other way of looking at it is to say that the particle is in all places at once (a superposition). How you think of this would depend on your particular interpretation of quantum mechanics.

 

[Borean]

According to the standard interpretation of QM, the outcome is fundamentally random and it is theoretically impossible to predict the outcome.

It's random, but that does not mean acausal.

We need to distinguish between determinacy and causality.

Indeterminacy does not mean acausality; it means indeterminacy.

If our ancestors millennia ago had no idea why the sun rose, no matter how hard they tried explaining it, that would not imply that the universe is acausal, but simply indeterminable.

Even if we couldn't determine why the sun rises today, we would still not have come to the conclusion of acausality; still only indeterminacy.

Of course, you can always suggest that maybe there is some more fundamental theory than QM with additional underlying variables, the knowledge of which would allow you to accurately predict the outcome of such experiments.

That is not necessary to affirm indeterminacy of isolated events in the universe.

Simply because, there is nothing that says any theory which has anything less than total information about the universe will ever be able to predict/determine the total amount of events in the universe, in isolation.

However, there are good reasons for believing that this is not the case (e.g., Bell's theorem), although it doesn't convince absolutely everyone.

But Bells theorem does nothing to suggest acausality over a fundamental indeterminacy.

 

[Borean]

QM is causal only to the extent that the wave function can accurately predict probabilities. But that is not what we normally understand as strict causality. Let's say we have a wave packet that represents a particle. We can predict how the shape and the position of the wave packet is going to evolve. But when we try to find the location of the particle, we can't say with certainty were we are going to find it. In this respect QM is acausal.

Again, that is indeterminacy, not acausality.

 

[Bob_for_short]

QM is acausal, as explained by the following experiments.

As soon as QM works with probabilities, it is clear that the events are occasional, spread, with non-trivial distributions. The physics task is to catch this property of nature and QM does it. The laws of ensembles (distributions) are described well, without hidden parameters.

 

[gel]

Indeterminacy does not mean acausality; it means indeterminacy.

Ok, I'm not really sure about the definitions here. Yes, QM is random. Is this the same thing as acausal? What caused the photon to pass through the polarizers rather than be absorbed? I don't know - QM just says that it is just a random event with given probabilities. If you consider that to be enough to constitute a cause, then it is causal.

 

[Bob_for_short]

Ok, I'm not really sure about the definitions here. Yes, QM is random. Is this the same thing as acausal? What caused the photon to pass through the polarizers rather than be absorbed? I don't know - QM just says that it is just a random event with given probabilities. If you consider that to be enough to constitute a cause, then it is causal.

The polariser presence means introducing an interaction (absorption). Any interaction has many channels, choices, paths, etc. No wonder sometimes photons are absorbed, sometimes not.

 

[gel]

I think Borean's only disagreement with my post was over the meaning of the word acausal, and that we agree about what QM is saying. Which is fine by me.

And I think Bob_for_short is agreeing with my post, but I'm not sure. QM tells us the probabilities for each outcome. If that is really the best that can be done (and I expect that it is), then we can't really expect anything more of it.

 

[gel]

The polariser presence means introducing an interaction (absorption). Any interaction has many channels, choices, paths, etc. No wonder sometimes photons are absorbed, sometimes not.

If the polariser is working properly, then
- a photon with polarization aligned with the polariser axis will pass through it.
- a photon with polarization at 90 deg to it will be absorbed.
That's all you need to be able to conclude that the probability is 50% in my experiment, regardless of how the photon actually interacts with the polarizer. Furthermore, it allows you to say that the probability is cos²(a) for an angle a between the polariser axes.

 

[Borean]

Can anyone tell me what is the cause of the phenomenon known as quantum tunnelling?

And additionally, the cause of the phenomenon of vacuum fluctuations or virtual particles?

 

[gel]

If our ancestors millennia ago had no idea why the sun rose, no matter how hard they tried explaining it, that would not imply that the universe is acausal, but simply indeterminable.

This doesn't seem right though. With QM we say that we don't know what the outcome of an experiment will be, not simply because we have no idea what is happening, but rather that it is believed to be impossible to do better than give a probability.
On the other hand, our ancestors had a pretty good idea of whether the sun would rise in the morning. It always did. They just were clueless as to why, and invented some rather fanciful explanations.

 

[Borean]

This doesn't seem right though. With QM we say that we don't know what the outcome of an experiment will be, not simply because we have no idea what is happening, but rather that it is believed to be impossible to do better than give a probability.
On the other hand, our ancestors had a pretty good idea of whether the sun would rise in the morning. It always did. They just were clueless as to why, and invented some rather fanciful explanations.

That wasn't my point really. My point was that even if we don't know or can't know the cause of a given thing, that in no way means there is no cause for given thing.

 

[gel]

That wasn't my point really. My point was that even if we don't know or can't know the cause of a given thing, that in no way means there is no cause for given thing.

We're in agreement then. We don't know whether the photon will pass through the polariser until after it has already passed through or been absorbed. Furthermore, I don't think we will ever be able to predict such things. That doesn't mean that there is no cause. I accept that there could be one, but if it can never be known then it is outside of what can be described by physics.

 

[Borean]

We're in agreement then. We don't know whether the photon will pass through the polariser until after it has already passed through or been absorbed. Furthermore, I don't think we will ever be able to predict such things. That doesn't mean that there is no cause. I accept that there could be one, but if it can never be known then it is outside of what can be described by physics.

Of course. My only point is that the indeterminacy need not be an inherent attribute of the observed thing, that is, "acausality"; it may as well lie in the observer.

Being a skeptic, I find it much more likely that the indeterminism is in the epistemic realm, not the natural realm. If we project it onto the natural realm, we are pretending to some kind of omniscience. Perverse, I say.

 

[gel]

Of course. My only point is that the indeterminacy need not be an inherent attribute of the observed thing, that is, "acausality"; it may as well lie in the observer.

I think that there is a difference between saying that the physics is causal/acausal and saying the same of the universe.
If there is a cause for the photon to pass through the polariser, then it is not included within standard QM. So, we can say that QM is acausal. There may or may not be a cause, but it isn't described by QM.
There are other models, such as the Bohmian interpretation for which there is a cause, and full knowledge of the initial state would allow you to decide the outcome beforehand. The problem with this is that it is generally assumed to satisfy a "quantum equilibrium" hypothesis, in which case the initial state can never be known exactly and you still can't predict the outcome any better than standard quantum mechanics. For this reason, it is often considered to be an interpretation rather than a new model. I personally wouldn't recommend the Bohmian interpretation, I mention it here just as an example.

 

[Borean]

I think that there is a difference between saying that the physics is causal/acausal and saying the same of the universe.
If there is a cause for the photon to pass through the polariser, then it is not included within standard QM. So, we can say that QM is acausal. There may or may not be a cause, but it isn't described by QM.

What do you mean? The Bohmian model isn't QM?

There are other models, such as the Bohmian interpretation for which there is a cause, and full knowledge of the initial state would allow you to decide the outcome beforehand. The problem with this is that it is generally assumed to satisfy a "quantum equilibrium" hypothesis, in which case the initial state can never be known exactly and you still can't predict the outcome any better than standard quantum mechanics. For this reason, it is often considered to be an interpretation rather than a new model. I personally wouldn't recommend the Bohmian interpretation, I mention it here just as an example.

It sounds like the Bohmian model exactly admits to agnosticism/indeterminacy in the human epistemic realm. A more likely explanation, in my opinion.

 

[Dale]

QM is not acausal. A causal system has a state which depends only on past or present values. An acausal system depends at least on some future values and possibly on past or present values. An anticausal system depends only on future values.

There is nothing in standard QM that implies any dependence on future values, so QM is causal. The randomness does not imply acausality unless it somehow depends on future values of the state.

 

[Borean]

Can anyone tell me what is the cause of the phenomenon known as quantum tunnelling?

And additionally, the cause of the phenomenon of vacuum fluctuations or virtual particles?

Could anyone answer this?

 

[Borean]

QM is not acausal. A causal system has a state which depends only on past or present values. An acausal system depends at least on some future values and possibly on past or present values. An anticausal system depends only on future values.

There is nothing in standard QM that implies any dependence on future values, so QM is causal. The randomness does not imply acausality unless it somehow depends on future values of the state.

Interesting insight.

 

[Pio2001]

Hello,
Here is the current state of affairs to my knowledge.

Let's use the definitions given by Borean : indeterminacy means that there are some events whose cause we don't know, and acausality that there are some events that have no cause at all.

It's been common practice to name "hidden variable" an unknown cause that would give birth to the values that quantum mechanics doesn't predict in the measurment process. So if hidden variables doesn't exist, then the universe is acausal.

In 1964, Bell proves that quantum mechanics is in contradiction with local hidden variables (hidden variables whose effect travels slower or at the speed of light).
In 1967, Kochen and Specker prove that no hidden variables can be carried by the measured system and bear all prédictions about their possible measured properties. If hidden variable exist, local or not, they depend at least party, on the measurment setup. They are contextual.
In 1969, Bell's theorem was generalized by Clauser, Horne, Shimony and Holt to contextual hidden variables.

In 1982, Alain Aspect's experiment shows that quantum mechanics predictions in Bell and CHSH theorems are fullfilled. Therefore if the universe is causal, then it is non-local.

The Kochen-Specker theorem have been tested too. The results of a recent experiment by Roos, Blatt et al have just been published in Nature. It confirms the inexistence of non-contextual hidden variables with great accuracy.

Another recent experiment by Gröblacher et al ruled out a whole class of non-local hidden variable interpretations : http://www.arxiv.org/PS_cache/arxiv/pdf/0704/0704.2529v1.pdf
I'm not sure which class exactly, but it seems to me that they are interpretations without entanglement.

So we're left with some philosophiocal problems. Experiments give us the choice between
-Leaving out causality : some events occur without any cause.
-Leaving out locality : the cause of an event can lie in its future.
-Leaving out realism : there is no reality. There is just information gathered by observers.

The Copenhagen interpretation makes both the first and third choices, as well as Rovelli's relational interpretation.
Cramer's transactional interpretation makes the second choice. As did Böhm-De Broglie's pilot wave interpretation (but they didn't know it before Bell's theorem).
Everett many-world's interpretations are usually not developped enough to get a precise position in this choice. JesseM, in this forum, used to describe an extension of Everett's interpretation that would get rid of these three problems ( but introducing the existence of many worlds of course) : https://www.physicsforums.com/showthread.php?t=206291#11