An argument against Bohmian mechanics?

[PeterDonis]

Because the gravitational action at a distance is not observable by any actual possible measurement of that distant place where it supposedly exerted its influence and therefore there is no way to use it to send information FTL.

I disagree if you are talking about what Newtonian mechanics predicts; Newtonian mechanics predicts that this should be possible. For example, according to Newtonian mechanics I should be able to wiggle the Sun and modulate the wiggle to transmit information like a radio does, and have you, on Earth, instantaneously detect the wiggle and decode the information like a radio receiver does, instead of having to wait the 8 minutes it would take for light to cover the distance. I agree that we don't actually observe this, but that's not because Newtonian mechanics does not predict it; it's because this prediction of Newtonian mechanics is incorrect.

 

[RockyMarciano]

I disagree if you are talking about what Newtonian mechanics predicts; Newtonian mechanics predicts that this should be possible. For example, according to Newtonian mechanics I should be able to wiggle the Sun and modulate the wiggle to transmit information like a radio does, and have you, on Earth, instantaneously detect the wiggle and decode the information like a radio receiver does, instead of having to wait the 8 minutes it would take for light to cover the distance. I agree that we don't actually observe this, but that's not because Newtonian mechanics does not predict it; it's because this prediction of Newtonian mechanics is incorrect.

But you can't modulate the wiggle if all motion is transmitted instantaneously, you are just deriving a prediction from the assumption of nonlocality. And in the process introducing something like the local field of EM.

 

[Mentz114]

But you can't modulate the wiggle if all motion is transmitted instantaneously...

I don't see why not. A movement could be picked up (in principle) by a gravitometer anywhere, instantaneously.

 

[Jilang]

I don't see why not. A movement could be picked up (in principle) by a gravitometer anywhere, instantaneously.

No, it can't. It's a wondrous thing also limited to c.

 

[RockyMarciano]

I don't see why not. A movement could be picked up (in principle) by a gravitometer anywhere, instantaneously.

To convey information(that is, "unexpected" signals) you need a finite speed of propagation of the "wiggles". Look up Fourier transform.

 

[Mentz114]

No, it can't. It's a wondrous thing also limited to c.

In this context we are assuming instantaneous propagation.

 

[Jilang]

Why?

 

[Mentz114]

To convey information(that is, "unexpected signals) you need a finite speed of propagation of the "wiggles". Look up Fourier transform.

Fourier what ? Never heard of it.

I'm reminded of the folksy saying 'time is there to stop everything happening at once'.

 

[Mentz114]

Why?

I forget.

[Edit]
I think it is to see if the assumption of instaneous signalling (IS) has any predictive power. But it seems that IS cannot exist. Which knocks back its predictive power by several megawatts.

 

[PeterDonis]

To convey information(that is, "unexpected" signals) you need a finite speed of propagation of the "wiggles". Look up Fourier transform.

What does that have to do with a finite propagation speed? You have a signal arriving at a detector, and the detected signal is a function of time, but that in no way requires that the signal was propagated at a finite speed from somewhere else. As long as the received signal as a function of time is correlated with the source signal as a function of time, you can transmit information. Fourier transform just converts the function of time to a function of frequency; but the "time" and "frequency" are in the detected signal, not in the propagation of anything. (At least, not if you are talking about signal processing; Fourier transforms do show up in other places, such as dealing with wave equations for propagation, but that doesn't mean that everywhere a Fourier transform is used there must be propagation at finite speed.)

 

[PeterDonis]

Why?

Do you mean, why are we assuming instantaneous propagation in this context? Because in this particular subthread we are talking about Newtonian gravity, which has instantaneous action at a distance.

 

[zonde]

they use "local" to mean deterministic hidden variables

Have not seen such a case. "Local" usually means that causality can't be faster than light. In QFT contexts "local" usually means that spacelike separated observables commute (or no FTL communication in simple words).

 

[RockyMarciano]

What does that have to do with a finite propagation speed? You have a signal arriving at a detector, and the detected signal is a function of time, but that in no way requires that the signal was propagated at a finite speed from somewhere else. As long as the received signal as a function of time is correlated with the source signal as a function of time, you can transmit information. Fourier transform just converts the function of time to a function of frequency; but the "time" and "frequency" are in the detected signal, not in the propagation of anything. (At least, not if you are talking about signal processing; Fourier transforms do show up in other places, such as dealing with wave equations for propagation, but that doesn't mean that everywhere a Fourier transform is used there must be propagation at finite speed.)

Since you brought up modulation I thought you knew that to modulate a waveform you have to vary its period in time and that is incompatible with instantaneity. I mentioned FT to give context about this. Now that you drop "wiggle modulation" you must bear in mind that assuming determinism requires independence of spatially separated points which in itself is form of locality assumption (the one used in the BI that classical mechanics follow) and that is broken in your latest example of instantaneous signal transmission, information at one point is intantaneously available at any other point. You'd lose the notion of spatial separation that is a condition of classical determinism like classical mechanics and hidden variables theories. And if the assumption of nonlocality automatically implied the prediction of ftl signal transmission if would be a prediction of Bohmian mechanics, but it isn't.

 

[RockyMarciano]

Have not seen such a case. "Local" usually means that causality can't be faster than light. In QFT contexts "local" usually means that spacelike separated observables commute (or no FTL communication in simple words).

Yes, I spent several posts insisting on just this. It's just that traditionally people used it as short for "local realism". You've seen it. You linked a post where it is used in this way

 

[stevendaryl]

Since you brought up modulation I thought you knew that to modulate a waveform you have to vary its period in time and that is incompatible with instantaneity.

I don't understand why you're saying this. Imagine taking a long stick and poking somebody with it. You've sent a message to that person. In Newtonian physics, if the stick were really perfectly rigid, the message would be sent instantaneously. In the actual world, when you move one end of the stick, there is a compression wave that travels down the stick, and the far end of the stick doesn't move until that wave reaches it. So in reality, using a stick to poke someone isn't instantaneous. But you can certainly imagine instantaneous communication.

 

[RockyMarciano]

I don't understand why you're saying this. Imagine taking a long stick and poking somebody with it. You've sent a message to that person. In Newtonian physics, if the stick were really perfectly rigid, the message would be sent instantaneously. In the actual world, when you move one end of the stick, there is a compression wave that travels down the stick, and the far end of the stick doesn't move until that wave reaches it. So in reality, using a stick to poke someone isn't instantaneous. But you can certainly imagine instantaneous communication.

Sure, but poking with an infinetely rigid rod has nothing to do with "wiggle modulation" and the part of my post you quoted is about modulation. In the second part of my post l explain why information can't be transmitted by poking for a deterministic theory. Instantaneous transmission of information would ruin independence of spatially separated regions, and if would violate BI which we know classical mechanics doesn't. So the non observable assumption of "action at a distance" of Newton's gravity must respect the existence of a distance, i.e. independence of spatially separated points, so again it doesn't predict ftl info sending.

 

[stevendaryl]

Sure, but poking with an infinetely rigid rod has nothing to do with "wiggle modulation" and the part of my post you quoted is about modulation. In the second part of my post l explain why information can't be transmitted by poking for a deterministic theory. Instantaneous transmission of information would ruin independence of spatially separated regions, and if would violate BI which we know classical mechanics doesn't. So the non observable assumption of "action at a distance" of Newton's gravity must respect the existence of a distance, i.e. independence of spatially separated points, so again it doesn't predict ftl info sending.

Okay, this just seems to me to be factually incorrect. It is not true that information can't be transmitted by poking for a deterministic theory. Whether classical mechanics violates Bell's inequalities is a matter of semantics. In deriving his inequality, he assumed that there was no way that information could be transmitted between the two measurement events in the EPR experiment. Why did he assume this? Because he assumed that no information could be transmitted faster than the speed of light. So he was ASSUMING no FTL information transmission when he derived his inequality. You can't turn around and use his inequality to prove that there is no FTL. That would be circular reasoning.

 

[RockyMarciano]

Okay, this just seems to me to be factually incorrect. It is not true that information can't be transmitted by poking for a deterministic theory. Whether classical mechanics violates Bell's inequalities is a matter of semantics. In deriving his inequality, he assumed that there was no way that information could be transmitted between the two measurement events in the EPR experiment. Why did he assume this? Because he assumed that no information could be transmitted faster than the speed of light. So he was ASSUMING no FTL information transmission when he derived his inequality. You can't turn around and use his inequality to prove that there is no FTL. That would be circular reasoning.

I'm insisting precisely that both locality and nonlocality are assumptios, not predictions for deterministic theories. Peterdonis said that nonlocality of Newtonian gravity predicted ftl info transmission and I explained why not(which doesn't mean that it predicts "no ftl transmission"). So I'm not saying either that the assumption of locality means that "no ftl transmission" is predicted or proved. I just showed that the opposite (ftl transmission) is not predicted by the assumption of nonlocality. For this the fact that classical mechanics doesn't violate the BI is not essential. You seem to have ignored the rest of the post's reasoning. So whether ftl info transmission is posible or not is just not a prediction of the theory.

 

[stevendaryl]

I'm insisting precisely that both locality and nonlocality are assumptios, not predictions for deterministic theories. Peterdonis said that nonlocality of Newtonian gravity predicted ftl info transmission and I explained why not.

I know. I'm just saying that you are wrong about that.

 

[RockyMarciano]

I know. I'm just saying that you are wrong about that.

why?

 

[stevendaryl]

why?

I can create a "tidal force" reader in the following way: I have two identical masses connected by a spring. For concreteness, let's pick a coordinate system where the two masses are lined up on the x-axis. We let the masses float in freefall. Then:

1. If the x-component of the gravitational force on the left mass is greater than that on the right mass, the spring will stretch.
2. If the x-component of the gravitational force on the left mass is less than that on the right mass, the spring will compress.

So a very precise measurement of the stretching of the mass will give an estimate of the variation of the gravitational field.

Now, suppose that the only gravitating mass is some supermassive rock that is a trillion miles away. Then Superman moving the rock in one direction or the other will instantaneously cause a change in the spring compression. So he can use "more compression/less compression" to encode 0 and 1 to transmit a message instantaneously.

 

[RockyMarciano]

I can create a "tidal force" reader in the following way: I have two identical masses connected by a spring. For concreteness, let's pick a coordinate system where the two masses are lined up on the x-axis. We let the masses float in freefall. Then:

1. If the x-component of the gravitational force on the left mass is greater than that on the right mass, the spring will stretch.
2. If the x-component of the gravitational force on the left mass is less than that on the right mass, the spring will compress.

So a very precise measurement of the stretching of the mass will give an estimate of the variation of the gravitational field.

Now, suppose that the only gravitating mass is some supermassive rock that is a trillion miles away. Then Superman moving the rock in one direction or the other will instantaneously cause a change in the spring compression. So he can use "more compression/less compression" to encode 0 and 1 to transmit a message instantaneously.

I already explained why this doesn't work for deterministic theories. Look if deterministic theories predicted ftl message transmission they could never be considered as local like Bell did. And I have already given an example where action at a distance doesn't imply the prediction of ftl signaling, namely Bohmian mechanics. Something similar to your setup could be imagined in BM and yet the fact remains it doesn't lead to a testable prediction.

 

[PeterDonis]

to modulate a waveform you have to vary its period in time and that is incompatible with instantaneity.

You have to vary the periodicity in time of whatever property of the source you are modulating, but that is a property of the source, not of something that propagates. And that is perfectly compatible with instantaneous "propagation" of something that depends on that property of the source and can be detected at a distance. An explicit example was even given by stevendaryl, a straightforward scenario based on Newtonian mechanics.

 

[PeterDonis]

I already explained why this doesn't work for deterministic theories.

Your "explanations" are simply not correct. Continuing to insist on them won't change that. There doesn't seem to be much point in continuing discussion, so this thread is closed.