Hello retrocausality!
July 10, 2017 6:22 AM Subscribe
Many people have heard of entanglement, an idea from quantum physics where two particles are connected in such a way that affecting one affects the other instantly, no matter where in the universe it is located.
(Entanglement probably deserves to be called more than an idea since it has been tested over and over again to plug loopholes that might show the particles are actually interacting on a local level in some way, in spite of what seem to be vast distances.)
But what if causality could be shown to work retrospectively somehow? Imagine if a particle could carry the effect backwards in time to when it was in very close proximity to its partner. No faster-than-light messages across the universe would be needed. Just backwards in time is all.
Say hello to retrocausality!
Previously.
A more easily digestible version of the article (with visual aids!), Space Time's "How the Quantum Eraser Rewrites the Past" episode.
posted by Talez at 6:38 AM on July 10, 2017 [4 favorites]
posted by Talez at 6:38 AM on July 10, 2017 [4 favorites]
MOATFI?
posted by lalochezia at 6:56 AM on July 10, 2017
posted by lalochezia at 6:56 AM on July 10, 2017
This Hua Price papaer looks rather accessable: https://arxiv.org/pdf/1510.06712.pdf
posted by bdc34 at 6:58 AM on July 10, 2017
posted by bdc34 at 6:58 AM on July 10, 2017
If this is true, then it is actually possible for things to happen for a reason.
posted by JohnFromGR at 7:05 AM on July 10, 2017
posted by JohnFromGR at 7:05 AM on July 10, 2017
> If this is true, then it is actually possible for things to happen for a reason.
And it is still possible for that reason to be "sometimes you're stupid and make bad decisions" ;)
posted by SNACKeR at 7:08 AM on July 10, 2017 [12 favorites]
And it is still possible for that reason to be "sometimes you're stupid and make bad decisions" ;)
posted by SNACKeR at 7:08 AM on July 10, 2017 [12 favorites]
If this is true, then it is actually possible for things to happen for a reason.
No. It doesn't work like that.
posted by Talez at 7:09 AM on July 10, 2017
No. It doesn't work like that.
posted by Talez at 7:09 AM on July 10, 2017
Please don't use quantum phenomena to understand or justify human behavior. That's not how any of it works at all.
posted by mysticreferee at 7:12 AM on July 10, 2017 [9 favorites]
posted by mysticreferee at 7:12 AM on July 10, 2017 [9 favorites]
How would this change Roko's Basilisk?
posted by acb at 7:16 AM on July 10, 2017 [2 favorites]
posted by acb at 7:16 AM on July 10, 2017 [2 favorites]
All of this is completely consistent with my fundamental belief about how the Universe works, which is that everything always happens everywhere and the only parts left over are the ones that happen not to have cancelled each other out.
posted by flabdablet at 7:18 AM on July 10, 2017 [8 favorites]
posted by flabdablet at 7:18 AM on July 10, 2017 [8 favorites]
How would this change Roko's Basilisk?
It doesn't. You're still an idiot if you believe in it.
posted by Talez at 7:19 AM on July 10, 2017 [12 favorites]
It doesn't. You're still an idiot if you believe in it.
posted by Talez at 7:19 AM on July 10, 2017 [12 favorites]
Talez: I was about to post that exact video, which pairs well with Space Time's "The Race to a Habitable Exoplanet - Time Warp Challenge" (and its solution, Space Time's "Superluminal Time Travel + Time Warp Challenge Answer"), which posit that FTL travel is *inherently* time travel. Alternately, you could look into the Picard Maneuver...
posted by mystyk at 7:19 AM on July 10, 2017 [1 favorite]
posted by mystyk at 7:19 AM on July 10, 2017 [1 favorite]
Also, causality is no more than a bunch of Just So stories made up after any given fact. It's got nothing fundamental to do with the actual workings of the Universe. Clue there is that giving a precise definition of what the word "causality" actually means has never actually been done.
posted by flabdablet at 7:22 AM on July 10, 2017
posted by flabdablet at 7:22 AM on July 10, 2017
> "It's got nothing fundamental to do with the actual workings of the Universe."
Wouldn't that imply that the universally consistent results of all observation on the subject, from casual to rigorously scientific, and therefore the contents of every known physical law, basically come down to some kind of massive coincidence?
I'm, let's call it, dubious.
posted by kyrademon at 7:32 AM on July 10, 2017
Wouldn't that imply that the universally consistent results of all observation on the subject, from casual to rigorously scientific, and therefore the contents of every known physical law, basically come down to some kind of massive coincidence?
I'm, let's call it, dubious.
posted by kyrademon at 7:32 AM on July 10, 2017
Also, causality is no more than a bunch of Just So stories made up after any given fact. It's got nothing fundamental to do with the actual workings of the Universe. Clue there is that giving a precise definition of what the word "causality" actually means has never actually been done.
Lorentz and Einstein would disagree. Without causality and the cosmic speed limit of causality we have no universe.
posted by Talez at 7:33 AM on July 10, 2017 [2 favorites]
Lorentz and Einstein would disagree. Without causality and the cosmic speed limit of causality we have no universe.
posted by Talez at 7:33 AM on July 10, 2017 [2 favorites]
Au contraire. Without causality and the cosmic speed limit we have no reasonable way to describe the universe.
posted by flabdablet at 7:39 AM on July 10, 2017
posted by flabdablet at 7:39 AM on July 10, 2017
tl;dr: E=mc²
For matter to exist requires energy proportional to mass times the square of the cosmic speed limit. As you increase c, the amount of energy needed for matter to exist increases. For you to have no causality would mean c is infinite, therefore the amount of energy needed for matter to exist is infinite which means no matter. We only have massleess particles whizzing about at infinite speed.
Causality is a very important thing to let us physically exist in these laws of reality.
posted by Talez at 7:40 AM on July 10, 2017 [1 favorite]
For matter to exist requires energy proportional to mass times the square of the cosmic speed limit. As you increase c, the amount of energy needed for matter to exist increases. For you to have no causality would mean c is infinite, therefore the amount of energy needed for matter to exist is infinite which means no matter. We only have massleess particles whizzing about at infinite speed.
Causality is a very important thing to let us physically exist in these laws of reality.
posted by Talez at 7:40 AM on July 10, 2017 [1 favorite]
> "How would this change Roko's Basilisk?"
I finally looked up what this is, having seen it referred to a few times on this site, and now I feel stupider for having that knowledge.
posted by kyrademon at 7:40 AM on July 10, 2017 [19 favorites]
I finally looked up what this is, having seen it referred to a few times on this site, and now I feel stupider for having that knowledge.
posted by kyrademon at 7:40 AM on July 10, 2017 [19 favorites]
There is this theory of the Moebius...
posted by Pogo_Fuzzybutt at 7:41 AM on July 10, 2017 [2 favorites]
posted by Pogo_Fuzzybutt at 7:41 AM on July 10, 2017 [2 favorites]
What always is left out from pop-sci discussion of this general topic, is that the so-called EPR paradox -- entanglement implying a violation of local causality -- is not a paradox (as it was intended to be by Einstein et al a criticism of QM) because you can never, ever validate this so-called action at a distance without doing so within the constraints of the light cone of the relationship you're validating. A changed state of an entangled particle is not knowable unless you know the state of both particles, and that knowledge is limited by local causality and GR. Whether it is across a vast distance, or alternatively, time, it doesn't matter because the altered state of an entangled particle is knowable only within the boundaries of local causality. It still sort of breaks your brain if you insist on interpreting the math according to some kind of naive realism -- but you really oughtn't be doing that. The math of QM is the reality of QM, not the translation into a narrative that's how we experience the world.
This is also what is wrong with the narrative interpretation of Heisenberg Uncertainty -- that observing something interferes with it or changes it. That's misleading. The wave function doesn't actually collapse in some naive realistic way but, rather, what is knowable about an interaction is defined and constrained by the math that allows or does not allow a wave function description. That this is a function of the observation is not evidence of a causal relationship between observation and measurement, it is inherent in what it means to know something about a quantum event.
That's my lay description -- I'm not a particle physicist. The quantum realm is weird and non-intuitive, but that's mostly the result of attempting to comprehend the math in a cognitive framework (intuitive narrative of physical interactions) that is ill-suited and misleading for this purpose.
That's my curmudgeonly lecture about popular science woo articles sensationalizing QM.
That said, the weirdness of QM is "real" insofar as our urge, our goal, is to find some intuitive narrative about the world and the philosophical implications of QM are very much worth contemplating, even though that went out of fashion within physics in the forties.
About the causality discussion -- causality doesn't have a rigorous definition. Ultimately, we only have correlation; "causality" is, in its essence, a philosophical assumption. But that has little relevance to the actual practice of science. In practice, causality is assumed to be knowable and, in fact, often known. And that's fine.
posted by Ivan Fyodorovich at 7:52 AM on July 10, 2017 [10 favorites]
This is also what is wrong with the narrative interpretation of Heisenberg Uncertainty -- that observing something interferes with it or changes it. That's misleading. The wave function doesn't actually collapse in some naive realistic way but, rather, what is knowable about an interaction is defined and constrained by the math that allows or does not allow a wave function description. That this is a function of the observation is not evidence of a causal relationship between observation and measurement, it is inherent in what it means to know something about a quantum event.
That's my lay description -- I'm not a particle physicist. The quantum realm is weird and non-intuitive, but that's mostly the result of attempting to comprehend the math in a cognitive framework (intuitive narrative of physical interactions) that is ill-suited and misleading for this purpose.
That's my curmudgeonly lecture about popular science woo articles sensationalizing QM.
That said, the weirdness of QM is "real" insofar as our urge, our goal, is to find some intuitive narrative about the world and the philosophical implications of QM are very much worth contemplating, even though that went out of fashion within physics in the forties.
About the causality discussion -- causality doesn't have a rigorous definition. Ultimately, we only have correlation; "causality" is, in its essence, a philosophical assumption. But that has little relevance to the actual practice of science. In practice, causality is assumed to be knowable and, in fact, often known. And that's fine.
posted by Ivan Fyodorovich at 7:52 AM on July 10, 2017 [10 favorites]
Causality is a very important thing to let us physically exist in these laws of reality.
Au contraire. Causality is a foundational concept within a set of laws that we have developed to describe our physical existence within this reality.
That being so, it's disappointing to find, on close and careful examination, just how sloppily defined the concept of causality actually is.
posted by flabdablet at 7:56 AM on July 10, 2017 [1 favorite]
Au contraire. Causality is a foundational concept within a set of laws that we have developed to describe our physical existence within this reality.
That being so, it's disappointing to find, on close and careful examination, just how sloppily defined the concept of causality actually is.
posted by flabdablet at 7:56 AM on July 10, 2017 [1 favorite]
In practice, causality is assumed to be knowable and, in fact, often known.
Indeed.
Ascribing some kind of fundamental necessity to it remains, in my view, unjustifiable.
posted by flabdablet at 7:59 AM on July 10, 2017
Indeed.
Ascribing some kind of fundamental necessity to it remains, in my view, unjustifiable.
posted by flabdablet at 7:59 AM on July 10, 2017
The wave function doesn't actually collapse in some naive realistic way but, rather, what is knowable about an interaction is defined and constrained by the math that allows or does not allow a wave function description.
That's close to the "Consistent Histories" interpretation. As far as I'm aware (admittedly, I've not been current for a couple decades now), it is consistent with the majority of QM theory and has not been experimentally disproven. It's important to realize however, that it is far from the only way to look at things.
posted by bonehead at 8:03 AM on July 10, 2017 [2 favorites]
That's close to the "Consistent Histories" interpretation. As far as I'm aware (admittedly, I've not been current for a couple decades now), it is consistent with the majority of QM theory and has not been experimentally disproven. It's important to realize however, that it is far from the only way to look at things.
posted by bonehead at 8:03 AM on July 10, 2017 [2 favorites]
> Meetup at the celebration.
several electrons from the future showed up but no one was interested in them so they left.
posted by toycamera at 8:18 AM on July 10, 2017
several electrons from the future showed up but no one was interested in them so they left.
posted by toycamera at 8:18 AM on July 10, 2017
Coincidentally (?), the universe provided today's xkcd.
posted by Huffy Puffy at 8:49 AM on July 10, 2017 [2 favorites]
posted by Huffy Puffy at 8:49 AM on July 10, 2017 [2 favorites]
So the reason why the cat got in the scanner lies in the future?
posted by ZeusHumms at 8:52 AM on July 10, 2017
posted by ZeusHumms at 8:52 AM on July 10, 2017
Excellent. I have always suspected that it is provincial of us to insist that causes must precede effects.
posted by thelonius at 8:53 AM on July 10, 2017 [1 favorite]
posted by thelonius at 8:53 AM on July 10, 2017 [1 favorite]
today's xkcd.
QM and GR are the sexy subjects, but the charms of fluid dynamics and three- (aqueous, NAPL, solid) and even four-phase (+air) systems are not to be denied.
posted by bonehead at 8:55 AM on July 10, 2017
QM and GR are the sexy subjects, but the charms of fluid dynamics and three- (aqueous, NAPL, solid) and even four-phase (+air) systems are not to be denied.
posted by bonehead at 8:55 AM on July 10, 2017
About the causality discussion -- causality doesn't have a rigorous definition. Ultimately, we only have correlation; "causality" is, in its essence, a philosophical assumption.
Yeah, it's a longstanding debate in metaphysics, wtf is causation anyway? It is very interesting, if you like that kind of thing. Unfortunately correlation alone isn't adequate: I see the same man every day on the train, but neither the train nor I are the cause of that. But you might counter they are parts of a complex cause, and then we're off....
But that has little relevance to the actual practice of science. In practice, causality is assumed to be knowable and, in fact, often known. And that's fine.
Totally agree. There's a form of scientism that kind of mistakes science for rationalist philosophy, where everything can be derived from first principles through the intellect, and that is of course not what science does. They also don't have to prove the existence of the external world before they can go to the lab.
posted by thelonius at 9:01 AM on July 10, 2017 [6 favorites]
Yeah, it's a longstanding debate in metaphysics, wtf is causation anyway? It is very interesting, if you like that kind of thing. Unfortunately correlation alone isn't adequate: I see the same man every day on the train, but neither the train nor I are the cause of that. But you might counter they are parts of a complex cause, and then we're off....
But that has little relevance to the actual practice of science. In practice, causality is assumed to be knowable and, in fact, often known. And that's fine.
Totally agree. There's a form of scientism that kind of mistakes science for rationalist philosophy, where everything can be derived from first principles through the intellect, and that is of course not what science does. They also don't have to prove the existence of the external world before they can go to the lab.
posted by thelonius at 9:01 AM on July 10, 2017 [6 favorites]
Next you people will be telling us that what happens beyond the event horizon doesn't actually stay beyond the event horizon.
posted by Talez at 9:14 AM on July 10, 2017 [2 favorites]
posted by Talez at 9:14 AM on July 10, 2017 [2 favorites]
Ivan Fyodorovich: Heisenberg Uncertainty […] That this is a function of the observation is not evidence of a causal relationship between observation and measurement, it is inherent in what it means to know something about a quantum event.
Is that right? (Probably, IF is much smarter than me.) I thought it meant that practically the measurement interfered with the particle, not that knowing position and momentum is theoretically impossible. Someone enlighten me before I embarrass myself in public.
posted by Johnny Wallflower at 9:14 AM on July 10, 2017
Is that right? (Probably, IF is much smarter than me.) I thought it meant that practically the measurement interfered with the particle, not that knowing position and momentum is theoretically impossible. Someone enlighten me before I embarrass myself in public.
posted by Johnny Wallflower at 9:14 AM on July 10, 2017
Please don't use quantum phenomena to understand or justify human behavior. That's not how any of it works at all.
Too late, I already have a book coming out, 8 TEDx talks scheduled, and a 3-year deal hosting a wellness show on OWN about "quantum spirituality".
posted by Sangermaine at 9:15 AM on July 10, 2017 [6 favorites]
Too late, I already have a book coming out, 8 TEDx talks scheduled, and a 3-year deal hosting a wellness show on OWN about "quantum spirituality".
posted by Sangermaine at 9:15 AM on July 10, 2017 [6 favorites]
I don't understand how this isn't the same as just reversing the signs on your time axis. It was my understanding that it was the universes generally increasing entropy that leads to the "illusion" of the "arrow of time".
posted by runcibleshaw at 9:17 AM on July 10, 2017
posted by runcibleshaw at 9:17 AM on July 10, 2017
I was going to read the actual paper but ran smack into the word "ontology" in the first line of the abstract and chickened out (at least for now). Dammit Jim, I'm a particle physicist, not a philosopher!
I am assuming that this new interpretation makes no new experimentally-testable predictions beyond conventional quantum mechanics, in which case the take-away message is "that's a cool way of looking at it, but whatevs".
posted by heatherlogan at 9:18 AM on July 10, 2017 [1 favorite]
I am assuming that this new interpretation makes no new experimentally-testable predictions beyond conventional quantum mechanics, in which case the take-away message is "that's a cool way of looking at it, but whatevs".
posted by heatherlogan at 9:18 AM on July 10, 2017 [1 favorite]
So in relativity, two points in space time are either "time-like" separated, or "space-like" separated. "Time-like" basically means that light moves fast enough that it could have gone from one point to another and "space-like" means it can't. So the point three feet to your right at this very instant is "space-like" separated from you, but the point 3 feet to the right whatever fraction of a second in the future that allows light to travel there is "time-like" separated. This is a fundamental distinction in relativity, because now matter what sort of crazy time-dilation type things are going on, time-like separated points stay time-like separated and space-like separated points stay space like separated, and space-like separated points never directly affect one another.
The whole "spooky action at a distance" and the EPR paradox is basically that quantum mechanics predicts certain correlations between particles emitted as pairs that can't be explained by any sort of "plan" the particles make beforehand that doesn't include "If you do this, then I'll do that." Which is a problem because the particles might be space-like separated at that point and those two points in space time should not be affecting each other at all. So the spooky action at a distance is the apparent interaction of two space-like separated points.
Of course one way to resolve this is to say that this paradox is the result of trying to impose classical intuitions onto a scale where things just don't work that way. Classical physics is just the large scale average of the quantum scale stuff, so quantum physics doesn't actually have to make sense classically as long as it all averages out.
Of course that sort of "shut up and calculate" feels really unsatisfying, so there's a strong temptation to reconcile what's going on another way. I think what this paper is proposing is that you can avoid the worry about space-like points interacting by assuming that causality can work both ways and you can have the future conditions of the particles affect the past.
Unfortunately, I'm thinking that this retro causality ends up being the same as spooky action at a distance. If you can send a message back in time at light speed three feet left, and then forward in time at light speed 3 more feet to the left, you've just instantaneously communicated between two points six feet apart. So I'm pretty sure FTL equals time travel works both ways: being able to travel both directions in time means moving faster than light.
posted by Zalzidrax at 9:22 AM on July 10, 2017 [2 favorites]
The whole "spooky action at a distance" and the EPR paradox is basically that quantum mechanics predicts certain correlations between particles emitted as pairs that can't be explained by any sort of "plan" the particles make beforehand that doesn't include "If you do this, then I'll do that." Which is a problem because the particles might be space-like separated at that point and those two points in space time should not be affecting each other at all. So the spooky action at a distance is the apparent interaction of two space-like separated points.
Of course one way to resolve this is to say that this paradox is the result of trying to impose classical intuitions onto a scale where things just don't work that way. Classical physics is just the large scale average of the quantum scale stuff, so quantum physics doesn't actually have to make sense classically as long as it all averages out.
Of course that sort of "shut up and calculate" feels really unsatisfying, so there's a strong temptation to reconcile what's going on another way. I think what this paper is proposing is that you can avoid the worry about space-like points interacting by assuming that causality can work both ways and you can have the future conditions of the particles affect the past.
Unfortunately, I'm thinking that this retro causality ends up being the same as spooky action at a distance. If you can send a message back in time at light speed three feet left, and then forward in time at light speed 3 more feet to the left, you've just instantaneously communicated between two points six feet apart. So I'm pretty sure FTL equals time travel works both ways: being able to travel both directions in time means moving faster than light.
posted by Zalzidrax at 9:22 AM on July 10, 2017 [2 favorites]
>I have always suspected that it is provincial of us to insist that causes must precede effects.
EFFECT, n. The second of two phenomena which always occur together in the same order. The first, called a Cause, is said to generate the other — which is no more sensible than it would be for one who has never seen a dog except in the pursuit of a rabbit to declare the rabbit the cause of the dog.
-- Ambrose Bierce
posted by lazycomputerkids at 9:22 AM on July 10, 2017 [4 favorites]
EFFECT, n. The second of two phenomena which always occur together in the same order. The first, called a Cause, is said to generate the other — which is no more sensible than it would be for one who has never seen a dog except in the pursuit of a rabbit to declare the rabbit the cause of the dog.
-- Ambrose Bierce
posted by lazycomputerkids at 9:22 AM on July 10, 2017 [4 favorites]
They also don't have to prove the existence of the external world before they can go to the lab
...and yet the world is plentifully supplied with writings from people who seem to proceed from the assumption that it wouldn't exist unless they had gone there, some of whom are respectable physicists.
Stephen Hawking, from A Brief History Of Time: "What is it that breathes fire into the equations and makes a universe for them to describe?"
I thought it meant that practically the measurement interfered with the particle, not that knowing position and momentum is theoretically impossible.
Main conceptual difficulty there, it seems to me, is expecting a subatomic particle to obey the same kind of object persistence rules that we learn as tiny children to apply to macroscopic objects.
Subatomic particles are just fundamentally unlike billiard balls. They're not so much a simple pointlike object as a propagating collection of tendencies. Without making a measurement, it's not really possible to know precisely what kind of thing you're proposing to take a measurement of.
posted by flabdablet at 9:23 AM on July 10, 2017 [2 favorites]
...and yet the world is plentifully supplied with writings from people who seem to proceed from the assumption that it wouldn't exist unless they had gone there, some of whom are respectable physicists.
Stephen Hawking, from A Brief History Of Time: "What is it that breathes fire into the equations and makes a universe for them to describe?"
I thought it meant that practically the measurement interfered with the particle, not that knowing position and momentum is theoretically impossible.
Main conceptual difficulty there, it seems to me, is expecting a subatomic particle to obey the same kind of object persistence rules that we learn as tiny children to apply to macroscopic objects.
Subatomic particles are just fundamentally unlike billiard balls. They're not so much a simple pointlike object as a propagating collection of tendencies. Without making a measurement, it's not really possible to know precisely what kind of thing you're proposing to take a measurement of.
posted by flabdablet at 9:23 AM on July 10, 2017 [2 favorites]
I've been really enjoying the Space Time videos; I try to absorb as much current physics by osmosis as I can, and every time I hit a brick wall there's always a bunch of other resources to help, so they're good jumping-off points even if they're often quite familiar ground. (Also a big fan of Leonard Susskind's lectures; his thought experiments with stuff like entanglement and black holes are endlessly fascinating.)
I can't remember which of the ST videos I saw the other day which discussed quantum tunnelling - a phenomenon which was superbly exotic when I first learned about it about thirty years ago and had about one actual use in technology, the tunnel diode, but now is literally everywhere in flash memory. The question was whether QT is superluminal - how long does it take to collapse the electron waveform on the far side of the wall? - and that's something I've wondered about before. (Oh, turns out that's the name of the video anyway.) Turns out it is, but as the video pointed out, any particle can appear anywhere the probabilities allow (ie, anywhere) and in a bog-standard beam of light you'll have photons which if observed will be in advance of where c says they should be. And that probably does break causality, but at a very tiny level.
posted by Devonian at 9:26 AM on July 10, 2017
I can't remember which of the ST videos I saw the other day which discussed quantum tunnelling - a phenomenon which was superbly exotic when I first learned about it about thirty years ago and had about one actual use in technology, the tunnel diode, but now is literally everywhere in flash memory. The question was whether QT is superluminal - how long does it take to collapse the electron waveform on the far side of the wall? - and that's something I've wondered about before. (Oh, turns out that's the name of the video anyway.) Turns out it is, but as the video pointed out, any particle can appear anywhere the probabilities allow (ie, anywhere) and in a bog-standard beam of light you'll have photons which if observed will be in advance of where c says they should be. And that probably does break causality, but at a very tiny level.
posted by Devonian at 9:26 AM on July 10, 2017
I thought it meant that practically the measurement interfered with the particle, not that knowing position and momentum is theoretically impossible.
Practically the measurement does indeed interfere with the particle. But also, knowing the position and momentum (with better than the minimum product of uncertainties) is indeed theoretically impossible [that's the epistomology], AND the particle even having a position and momentum that is "less smeared" than the minimum product of uncertainties is also impossible [that's the ontology - ok I used the word].
One way to think about it is to make the analogy to sound waves. The pitch of the sound is controlled by the frequency (or wavelength) of the wave -- that's analogous to the momentum in quantum mechanics. If I play middle C for 1 second, the pressure wave produced will travel outward from my [pretend] piano and occupy a certain length of space at any particular moment -- that's analogous to the position in quantum mechanics. Now imagine playing that note for a shorter and shorter time (narrowing down the "length" in space of the pressure wave). If you try to make the sound pulse shorter than the wavelength that corresponds to middle C, it's no longer going to be middle C. The fancy way of representing this mathematically is to decompose the resulting waveform into Fourier modes: the sound pulse starts getting other frequency modes mixed into it in order to form that non-sinusoidal waveform. That's analogous to the spread in momentum states that comes in when you try to localize a particle in quantum mechanics.
posted by heatherlogan at 9:29 AM on July 10, 2017 [5 favorites]
Practically the measurement does indeed interfere with the particle. But also, knowing the position and momentum (with better than the minimum product of uncertainties) is indeed theoretically impossible [that's the epistomology], AND the particle even having a position and momentum that is "less smeared" than the minimum product of uncertainties is also impossible [that's the ontology - ok I used the word].
One way to think about it is to make the analogy to sound waves. The pitch of the sound is controlled by the frequency (or wavelength) of the wave -- that's analogous to the momentum in quantum mechanics. If I play middle C for 1 second, the pressure wave produced will travel outward from my [pretend] piano and occupy a certain length of space at any particular moment -- that's analogous to the position in quantum mechanics. Now imagine playing that note for a shorter and shorter time (narrowing down the "length" in space of the pressure wave). If you try to make the sound pulse shorter than the wavelength that corresponds to middle C, it's no longer going to be middle C. The fancy way of representing this mathematically is to decompose the resulting waveform into Fourier modes: the sound pulse starts getting other frequency modes mixed into it in order to form that non-sinusoidal waveform. That's analogous to the spread in momentum states that comes in when you try to localize a particle in quantum mechanics.
posted by heatherlogan at 9:29 AM on July 10, 2017 [5 favorites]
I found the non-paywalled preprint version of the Leifer & Pusey paper: arXiv:1607.07871.
posted by heatherlogan at 9:40 AM on July 10, 2017 [2 favorites]
posted by heatherlogan at 9:40 AM on July 10, 2017 [2 favorites]
I had an early awareness of Einstein because we share a birthday and I gradually developed two knee-jerk musings upon any reference to him. The first is to recite the famous equation by all its terms (something he taught me)-- "solving" for mass and the constant. The second is recalling my failure to discover his middle name, or a reference confirming he doesn't have one.
And I mutter to myself in times of despair: Not only does God play dice, they're loaded.
posted by lazycomputerkids at 9:44 AM on July 10, 2017
And I mutter to myself in times of despair: Not only does God play dice, they're loaded.
posted by lazycomputerkids at 9:44 AM on July 10, 2017
space-like separated points never directly affect one another
...in the sense that you can't create a completely arbitrary event and guarantee that it will correlate with any other event that's space-like separated from it.
But what you can do is arrange for the creation of arbitrary sets of space-like-separated events that correlate with each other, and then act all surprised when those correlations turn out to work.
"Spooky" action at a distance is just science reporters beating up the consequences of observations conceptually equivalent to noting that seeing heads on one side of a coin guarantees that there will instantaneously be tails on the other, even if the coin has been sliced in two and the other half sent to Mars.
Really all that ever makes this rather pedestrian fact seem "spooky" is that our minds are simply not equipped to understand subatomic particles intuitively. We insist on making a conceptual separation between what we know about the structure we think of as a pair of entangled particles and what the posited particles are "actually doing", when in fact we don't even have a good justification for believing that the latter idea actually means anything at all in the absence of a measurement.
Not only does God play dice, they're loaded.
...which is just as well for us; quantum mechanics is actually a formalism for describing exactly how they're loaded.
posted by flabdablet at 9:48 AM on July 10, 2017 [3 favorites]
...in the sense that you can't create a completely arbitrary event and guarantee that it will correlate with any other event that's space-like separated from it.
But what you can do is arrange for the creation of arbitrary sets of space-like-separated events that correlate with each other, and then act all surprised when those correlations turn out to work.
"Spooky" action at a distance is just science reporters beating up the consequences of observations conceptually equivalent to noting that seeing heads on one side of a coin guarantees that there will instantaneously be tails on the other, even if the coin has been sliced in two and the other half sent to Mars.
Really all that ever makes this rather pedestrian fact seem "spooky" is that our minds are simply not equipped to understand subatomic particles intuitively. We insist on making a conceptual separation between what we know about the structure we think of as a pair of entangled particles and what the posited particles are "actually doing", when in fact we don't even have a good justification for believing that the latter idea actually means anything at all in the absence of a measurement.
Not only does God play dice, they're loaded.
...which is just as well for us; quantum mechanics is actually a formalism for describing exactly how they're loaded.
posted by flabdablet at 9:48 AM on July 10, 2017 [3 favorites]
This stuff is also why you can never have a truly monochromatic light beam, because - depending on how you choose to interpret it - you simply can't constrain frequency infinitely when observing photons. Or because you have to turn the thing on at some point, and Fourier gets you. Or you try plugging a zero-bandwidth signal into Shannon and see what happens. Or just 'that's not how photons do'. Or 'that's not how field perturbations do', seeing as particles and packets are all a bit old hat now.
This is one of the joys of QM; it forces you to realise that your mental models, even ones that seem pure, simple and obvious, do not match reality, and there are unspoken assumptions behind much of what you might read - everyone talks about monochromatic light sources but there's no such beast. It reminds me of the old gotcha exam question where you have a thin sheet of a conductor of a particular areal resistivity, of such-and-such a shape and size, and you're asked to calculate the resistance between two points within it. And the answer, which you can arrive at by sweat and doubt or by realisation, is infinite - a point has no dimensions, and you can't put any electricity through something that doesn't exist outside maths.
posted by Devonian at 9:53 AM on July 10, 2017 [2 favorites]
This is one of the joys of QM; it forces you to realise that your mental models, even ones that seem pure, simple and obvious, do not match reality, and there are unspoken assumptions behind much of what you might read - everyone talks about monochromatic light sources but there's no such beast. It reminds me of the old gotcha exam question where you have a thin sheet of a conductor of a particular areal resistivity, of such-and-such a shape and size, and you're asked to calculate the resistance between two points within it. And the answer, which you can arrive at by sweat and doubt or by realisation, is infinite - a point has no dimensions, and you can't put any electricity through something that doesn't exist outside maths.
posted by Devonian at 9:53 AM on July 10, 2017 [2 favorites]
there are unspoken assumptions behind much of what you might read
One of my own biggest "Aha!" moments involved noticing the twin unspoken assumptions that (a) experiments are ever truly in every conceivable detail repeatable and (b) we will always know what the relevant details actually are.
Debugging device drivers can teach a person many useful things.
posted by flabdablet at 10:01 AM on July 10, 2017 [1 favorite]
One of my own biggest "Aha!" moments involved noticing the twin unspoken assumptions that (a) experiments are ever truly in every conceivable detail repeatable and (b) we will always know what the relevant details actually are.
Debugging device drivers can teach a person many useful things.
posted by flabdablet at 10:01 AM on July 10, 2017 [1 favorite]
Please don't use quantum phenomena to understand or justify human behavior. That's not how any of it works at all.
Well, you can blame Roger Penrose, who published a theory that quantum effects give rise to consciousness in the brain, for giving a patina of respectability to that industry. I haven't read his book and I don't know if he argues that it's not possible to have a materialist theory of consciousness without this move, or what. He's a distinguished scientist but it kind of sounded like a side trip into crankery, to me. God of the gaps stuff. But maybe he's right, and the well is just poisoned.
posted by thelonius at 10:01 AM on July 10, 2017
Well, you can blame Roger Penrose, who published a theory that quantum effects give rise to consciousness in the brain, for giving a patina of respectability to that industry. I haven't read his book and I don't know if he argues that it's not possible to have a materialist theory of consciousness without this move, or what. He's a distinguished scientist but it kind of sounded like a side trip into crankery, to me. God of the gaps stuff. But maybe he's right, and the well is just poisoned.
posted by thelonius at 10:01 AM on July 10, 2017
a theory that quantum effects give rise to consciousness in the brain
Hello retrocausality!
posted by flabdablet at 10:03 AM on July 10, 2017 [1 favorite]
Hello retrocausality!
posted by flabdablet at 10:03 AM on July 10, 2017 [1 favorite]
I finally looked up what this is, having seen it referred to a few times on this site, and now I feel stupider for having that knowledge.
Your eternal torment of knowing such a stupid thing exists has begun.
posted by solotoro at 10:12 AM on July 10, 2017
Your eternal torment of knowing such a stupid thing exists has begun.
posted by solotoro at 10:12 AM on July 10, 2017
Oh, so this is where deja vu comes from? Believe you me, the past is avoiding us like the plague. If you feel the tingle, follow it to the emergency exit.
posted by Oyéah at 10:13 AM on July 10, 2017 [1 favorite]
posted by Oyéah at 10:13 AM on July 10, 2017 [1 favorite]
you can blame Roger Penrose, who published a theory that quantum effects give rise to consciousness in the brain
I read that when I was an undergrad! The Emperor's New Mind, I think it was called. The one good thing about it is that I learned what a universal Turing machine is. Penrose's thesis is that you can have a "program" or math problem or whatever a Turing machine runs, where you don't know whether it has a halting state or not unless you run it until it stops (or an infinitely long time, whichever comes first). Somehow we humans are supposed to be smarter than that because we can stop computation or something. [Whereas I think that this is just due to the fact that we are embodied cogitators -- we get bored or hungry or have to pee and put the problem down.]
Considering that chemistry can't exist without quantum effects, I think the statement that quantum effects give rise to consciousness in the brain is trivially true, and largely uninteresting.
posted by heatherlogan at 10:16 AM on July 10, 2017 [4 favorites]
I read that when I was an undergrad! The Emperor's New Mind, I think it was called. The one good thing about it is that I learned what a universal Turing machine is. Penrose's thesis is that you can have a "program" or math problem or whatever a Turing machine runs, where you don't know whether it has a halting state or not unless you run it until it stops (or an infinitely long time, whichever comes first). Somehow we humans are supposed to be smarter than that because we can stop computation or something. [Whereas I think that this is just due to the fact that we are embodied cogitators -- we get bored or hungry or have to pee and put the problem down.]
Considering that chemistry can't exist without quantum effects, I think the statement that quantum effects give rise to consciousness in the brain is trivially true, and largely uninteresting.
posted by heatherlogan at 10:16 AM on July 10, 2017 [4 favorites]
Penrose is also one of those on board the "breathes fire into the equations"/"mathematics exists in a separate realm from physical reality" train.
posted by flabdablet at 10:24 AM on July 10, 2017
posted by flabdablet at 10:24 AM on July 10, 2017
Also, if there's anybody out there who thinks universal Turing machines are cool but hasn't read GEB: strongly recommend you do so.
posted by flabdablet at 10:27 AM on July 10, 2017
posted by flabdablet at 10:27 AM on July 10, 2017
"Spooky" action at a distance is just science reporters beating up the consequences of observations conceptually equivalent to noting that seeing heads on one side of a coin guarantees that there will instantaneously be tails on the other, even if the coin has been sliced in two and the other half sent to Mars.
It's not though, which is exactly what bothers scientists, but layman's explanation usually don't go into enough detail to explain why. It requires at least some trigonometry and math.
So you have two entangled particles flying off, so they have a conserved, quantized property that you're measuring, like spin. When you detect it, you have to pick a direction you're detecting the spin in. For the simplest particles, electrons, the only values you can get are +1/2 (up) in that direction or -1/2 (down) in that direction. So if your two detectors are exactly aligned, one entangled electron has to register as up, the other has to register as down. Great, so far that fits the coin analogy.
The tricky part comes when the detectors are slightly off from each other. So when you rotate one of the detectors, you get a projection of the wave function from one to the other that's less than 1, but in quantum physics, the probability is going to be the square of the wave function. So let's call the first orientation A, and rotate one of the detectors 30 degrees to orientation B. You'll project sqrt(3)/2 of your wave function at B onto A, so when you go to probability you will end up with sqrt(3)/2 squared or 3/4 of the events being (anti)correlated. Three quarters of the time, you get up/down pairs, the rest is random.
Okay, so say you rotate the second detector 30 degrees further, to orientation C. Now the projection between C and A is only 1/2, so the correlation is 1/4. One quarter of the time you will get up/down pairs, the rest will be random. But if you had the first detector at orientation B with the second at C, then they'd only be 30 degrees separated and you'd have 3/4 of the particles correlating as up/down pairs again.
So this, if you think about it enough, is a problem. Classically, if these particles had picked out what they were going to do beforehand - It's just... weird.like two sides of a coin - then if only 1/4 of the particles had no correlation between A and B and 1/4 of the particles had no correlation between B and C, there's no way to end up with 3/4 of the particles not correlating between A and C. If only 1/4 of the particles "change their minds" going from A to B, and another 1/4 "change their minds" from B to C, then you'd still have at least half the particles registering as down when the other is up once you've gotten to orientation C. That's not what quantum physics predicts, and is not what happens. But, due to quantum physics, you can only measure each particle once and that changes their spin and disentangles them, so you don't actually run into any paradox in practice. It is physically impossible to measure the spin of a pair of particles at all three orientations A, B, and C all at once since there's only two particles to measure. But from a classical perspective, where everything has a definite state all the time, it makes no sense.
I mean there are a bunch of possible loopholes here, notably that a lot of the "uncorrelated" particle detections can still randomly happen to be up/down pairs just by the randomness of things, so doing the demonstration "right" is a bit more tricky. However, that sort of detail is a bit beyond the scope of a mefi thread.
posted by Zalzidrax at 10:35 AM on July 10, 2017 [9 favorites]
It's not though, which is exactly what bothers scientists, but layman's explanation usually don't go into enough detail to explain why. It requires at least some trigonometry and math.
So you have two entangled particles flying off, so they have a conserved, quantized property that you're measuring, like spin. When you detect it, you have to pick a direction you're detecting the spin in. For the simplest particles, electrons, the only values you can get are +1/2 (up) in that direction or -1/2 (down) in that direction. So if your two detectors are exactly aligned, one entangled electron has to register as up, the other has to register as down. Great, so far that fits the coin analogy.
The tricky part comes when the detectors are slightly off from each other. So when you rotate one of the detectors, you get a projection of the wave function from one to the other that's less than 1, but in quantum physics, the probability is going to be the square of the wave function. So let's call the first orientation A, and rotate one of the detectors 30 degrees to orientation B. You'll project sqrt(3)/2 of your wave function at B onto A, so when you go to probability you will end up with sqrt(3)/2 squared or 3/4 of the events being (anti)correlated. Three quarters of the time, you get up/down pairs, the rest is random.
Okay, so say you rotate the second detector 30 degrees further, to orientation C. Now the projection between C and A is only 1/2, so the correlation is 1/4. One quarter of the time you will get up/down pairs, the rest will be random. But if you had the first detector at orientation B with the second at C, then they'd only be 30 degrees separated and you'd have 3/4 of the particles correlating as up/down pairs again.
So this, if you think about it enough, is a problem. Classically, if these particles had picked out what they were going to do beforehand - It's just... weird.like two sides of a coin - then if only 1/4 of the particles had no correlation between A and B and 1/4 of the particles had no correlation between B and C, there's no way to end up with 3/4 of the particles not correlating between A and C. If only 1/4 of the particles "change their minds" going from A to B, and another 1/4 "change their minds" from B to C, then you'd still have at least half the particles registering as down when the other is up once you've gotten to orientation C. That's not what quantum physics predicts, and is not what happens. But, due to quantum physics, you can only measure each particle once and that changes their spin and disentangles them, so you don't actually run into any paradox in practice. It is physically impossible to measure the spin of a pair of particles at all three orientations A, B, and C all at once since there's only two particles to measure. But from a classical perspective, where everything has a definite state all the time, it makes no sense.
I mean there are a bunch of possible loopholes here, notably that a lot of the "uncorrelated" particle detections can still randomly happen to be up/down pairs just by the randomness of things, so doing the demonstration "right" is a bit more tricky. However, that sort of detail is a bit beyond the scope of a mefi thread.
posted by Zalzidrax at 10:35 AM on July 10, 2017 [9 favorites]
Classically, if these particles had picked out what they were going to do beforehand - It's just... weird.
That's exactly the same kind of weirdness, though, as you get from measuring a direction-sensitive attribute of a single particle and then measuring again after tilting the detector.
The only way it actually turns spooky weird is if you absolutely insist on thinking of the space-like-separated halves of the entangled pair as in any way comprehensible other than as a complete system. If you start believing that an entangled pair ought to be free to behave in ways available to two non-entangled particles, you fail to notice that it was the old guy from the fishing shack in a scary mask all along.
It is physically impossible to measure the spin of a pair of particles at all three orientations A, B, and C all at once since there's only two particles to measure.
There really, really aren't two particles, only the one entangled-pair system.
posted by flabdablet at 10:52 AM on July 10, 2017 [4 favorites]
That's exactly the same kind of weirdness, though, as you get from measuring a direction-sensitive attribute of a single particle and then measuring again after tilting the detector.
The only way it actually turns spooky weird is if you absolutely insist on thinking of the space-like-separated halves of the entangled pair as in any way comprehensible other than as a complete system. If you start believing that an entangled pair ought to be free to behave in ways available to two non-entangled particles, you fail to notice that it was the old guy from the fishing shack in a scary mask all along.
It is physically impossible to measure the spin of a pair of particles at all three orientations A, B, and C all at once since there's only two particles to measure.
There really, really aren't two particles, only the one entangled-pair system.
posted by flabdablet at 10:52 AM on July 10, 2017 [4 favorites]
Retrocausality, according to this view, would then amount to a crutch for intuition by allowing the entangled-pair system to be thought of as equivalent to a single particle that "travels" backward in time from the first measurement event to the entanglement event and then forward to the second measurement event.
Personally I'm happier with leaving causality as ill-defined as it actually is with reference to stuff assumed to be happening inside a quantum system, entangled or not. We have the wave function for the system, we have the measurement probability predictions derived from that, the probabilities check out when tested, it's all good.
I have absolutely no idea what causal mechanisms could possibly operate inside an entangled pair in order to "make" the measurements come out the way they do, and have seen no evidence that anybody else does either; a causal narrative, then, seems to me to inapplicable to understanding this behavior but - crucially - no more so than it is to the weird but non-spooky results of successive measurements on single particles.
Doing sufficient violence to the notion of causality as to make time-reversed causal particle histories a thing seems to me unlikely to leave it fit for purpose. I'd rather just have causality altogether inapplicable to certain kinds of analysis than turn it into even more of a Just So story than it already is.
posted by flabdablet at 11:12 AM on July 10, 2017 [1 favorite]
Personally I'm happier with leaving causality as ill-defined as it actually is with reference to stuff assumed to be happening inside a quantum system, entangled or not. We have the wave function for the system, we have the measurement probability predictions derived from that, the probabilities check out when tested, it's all good.
I have absolutely no idea what causal mechanisms could possibly operate inside an entangled pair in order to "make" the measurements come out the way they do, and have seen no evidence that anybody else does either; a causal narrative, then, seems to me to inapplicable to understanding this behavior but - crucially - no more so than it is to the weird but non-spooky results of successive measurements on single particles.
Doing sufficient violence to the notion of causality as to make time-reversed causal particle histories a thing seems to me unlikely to leave it fit for purpose. I'd rather just have causality altogether inapplicable to certain kinds of analysis than turn it into even more of a Just So story than it already is.
posted by flabdablet at 11:12 AM on July 10, 2017 [1 favorite]
what causal mechanisms could possibly operate inside an entangled pair
Quantum evolution as described by Schrodinger's equation is entirely causal. The EPR postulate ingredients that have to be given up are either locality or "reality" ("reality" meaning that each particle is in some definite state, i.e., no superpositions).
posted by heatherlogan at 11:29 AM on July 10, 2017
Quantum evolution as described by Schrodinger's equation is entirely causal. The EPR postulate ingredients that have to be given up are either locality or "reality" ("reality" meaning that each particle is in some definite state, i.e., no superpositions).
posted by heatherlogan at 11:29 AM on July 10, 2017
And if entanglement of particles isn't quite cutting it for you on the crazification spectrum, how about quantum entanglement of quasiparticles?
I can just about cope with holes in semiconductors behaving like particles, because even at a classical level that makes sense. Phonons, OK; you can look at stuff like eddies in a river and sure, they clearly have existence and interactions that make more sense if you think of them as things, rather than not-things, that follow similar patterns to other actual things. Electrons splitting into spinons, orbitons and for all I know, Mysterons - why not, if they fancy it. All just probabilistic perturbations of patterns in fields.
But you can entangle the not-things?
At this point, I tend to start asking myself 'does the stuff produce the rules, or the rules produce the stuff', and considering a career change to acid-taster for the Dali Lama.
posted by Devonian at 12:15 PM on July 10, 2017 [4 favorites]
I can just about cope with holes in semiconductors behaving like particles, because even at a classical level that makes sense. Phonons, OK; you can look at stuff like eddies in a river and sure, they clearly have existence and interactions that make more sense if you think of them as things, rather than not-things, that follow similar patterns to other actual things. Electrons splitting into spinons, orbitons and for all I know, Mysterons - why not, if they fancy it. All just probabilistic perturbations of patterns in fields.
But you can entangle the not-things?
At this point, I tend to start asking myself 'does the stuff produce the rules, or the rules produce the stuff', and considering a career change to acid-taster for the Dali Lama.
posted by Devonian at 12:15 PM on July 10, 2017 [4 favorites]
Penrose is also one of those on board the "breathes fire into the equations"/"mathematics exists in a separate realm from physical reality" train.
I have a soft spot in my heart (head?) for Platonism, myself.
posted by thelonius at 12:23 PM on July 10, 2017
I have a soft spot in my heart (head?) for Platonism, myself.
posted by thelonius at 12:23 PM on July 10, 2017
Ascribing some kind of fundamental necessity to it remains, in my view, unjustifiable.
Could you say what you mean by this? I'm not sure what "fundamental necessity" is supposed to mean or what it might mean to ascribe it to causation. Are you denying that causation (when it obtains) is a necessary connection between events? Or that causation is necessary for doing science? Or something else?
posted by Jonathan Livengood at 4:14 PM on July 10, 2017
Could you say what you mean by this? I'm not sure what "fundamental necessity" is supposed to mean or what it might mean to ascribe it to causation. Are you denying that causation (when it obtains) is a necessary connection between events? Or that causation is necessary for doing science? Or something else?
posted by Jonathan Livengood at 4:14 PM on July 10, 2017
Main conceptual difficulty there, it seems to me, is expecting a subatomic particle to obey the same kind of object persistence rules that we learn as tiny children to apply to macroscopic objects.
Preach it! That is well-said. Also, the speculative thinkers fond of explaining broad features of reality or experience by analogy to QM concepts are thinking as metaphysical atomists, who think we can explain reality by grinding down to the most fundamental unit of it. But that assumption, arguing from the subatomic realm, seems even less warranted than it was when they thought they had found the billiard balls.
posted by thelonius at 5:34 PM on July 10, 2017
Preach it! That is well-said. Also, the speculative thinkers fond of explaining broad features of reality or experience by analogy to QM concepts are thinking as metaphysical atomists, who think we can explain reality by grinding down to the most fundamental unit of it. But that assumption, arguing from the subatomic realm, seems even less warranted than it was when they thought they had found the billiard balls.
posted by thelonius at 5:34 PM on July 10, 2017
Quantum evolution as described by Schrodinger's equation is entirely causal
...and does not predict the outcomes of individual measurements, only the probabilities of those outcomes.
The point I'm trying to get at is that the sense of spookiness attached to the actual measurement results that have been shown to apply to entangled systems rests on an intuitive conception that Schrodinger's equation is not telling us the whole causal story; that instead it's in some way a summary or approximation for a theory of some classically causal mechanism hidden "inside" a quantum system that would provide 100% certainty for predictions of measurement outcomes if only we had access to it ("reality") - and where "inside" carries the customary implication that what is inside the system is in principle separable from influences outside ("locality").
As far as I know, this intuitive conception is just wrong.
I have no trouble at all with setting aside the "reality" assumption, since it seems to me that fundamental particles (especially "bare" particles) are more like mathematical conveniences than tiny little versions of anything one could hold in one's hand.
I am also not bothered by setting aside the "locality" assumption, since it seems at least plausible to me that the entire universe may well be best thought of as a single entangled system that never actually decoheres; what with being everything and all, it has no environment and there is therefore nowhere for information to escape to.
I'm not sure what "fundamental necessity" is supposed to mean or what it might mean to ascribe it to causation.
I'm saying that the most fundamental feature of any sound theory of how stuff actually works must rest squarely on the observation that something is happening, and that saying anything more precise than that about what is happening (or has, or will) requires making distinctions i.e. a conceptual carving-up of what's happening into distinguishable parts.
I'm saying that we have all been performing just such a carving-up since before we learned to speak. I think the fundamental feature of every person's personal physics is the distinction between "me" and "everything else", and that that's a distinction made so early that its influence over the rest of our understanding is inescapable unless we choose not to operate our brains in Safe Mode.
In particular, it seems to me that the idea of the universe as being composed of conceptually separable bodies rests squarely on the initial conceptual separation we've all made between our own bodies - that over which we can evidently exert control - and everything else, which we can't.
It also seems to me that the fundamental division of experience into past (that which we remember) and future (that which can surprise us) would have to happen very very early on as well.
Are you denying that causation (when it obtains) is a necessary connection between events?
Causation, loosely speaking, is the general principle that if this-kind-of-thing happens, that-kind-of-thing will surely follow. Before needing causation in order to give oneself an account of the world, one must invent not only distinctions but classification: the knowledge that this thing is one of those things. So it seems to me that causation is a principle that connects our internal models of events.
I don't believe that causation connects things in and of themselves in such a way as to rule out the possibility of events that have no cause even in principle, because I'm not convinced that the distinctions and classifications we need to make in order to identify events are in fact anything more than conveniences and conventions.
Or that causation is necessary for doing science?
Causation is undoubtedly a very useful principle, and works extremely well for acquiring and communicating understanding of a vast array of observed behaviour. But what I'm suggesting is that it's not the only useful conceptual tool applicable to understanding the correlations we observe in and around us, and that the dissatisfaction people experience on finding out that one of our best theoretical frameworks provides "mere" probability predictions rather than 100% repeatable if-this-then-that is actually completely unwarranted.
Because when it comes right down to it, the predictions made by every physical theory are probabilistic. There is no way to guarantee with 100% certainty that the conditions upon which you base the non-probabilistic predictions provided by a classical theory actually apply to any given piece of reality whose behaviour you are attempting to predict.
The beauty of quantum theory, to me, is in the way it wraps up that very uncertainty, incorporates it inside the theory itself, and turns it into probabilities good enough to do engineering with. It's an astounding intellectual achievement, and I strongly doubt I'll see it bettered before I die.
posted by flabdablet at 5:43 PM on July 10, 2017 [3 favorites]
...and does not predict the outcomes of individual measurements, only the probabilities of those outcomes.
The point I'm trying to get at is that the sense of spookiness attached to the actual measurement results that have been shown to apply to entangled systems rests on an intuitive conception that Schrodinger's equation is not telling us the whole causal story; that instead it's in some way a summary or approximation for a theory of some classically causal mechanism hidden "inside" a quantum system that would provide 100% certainty for predictions of measurement outcomes if only we had access to it ("reality") - and where "inside" carries the customary implication that what is inside the system is in principle separable from influences outside ("locality").
As far as I know, this intuitive conception is just wrong.
I have no trouble at all with setting aside the "reality" assumption, since it seems to me that fundamental particles (especially "bare" particles) are more like mathematical conveniences than tiny little versions of anything one could hold in one's hand.
I am also not bothered by setting aside the "locality" assumption, since it seems at least plausible to me that the entire universe may well be best thought of as a single entangled system that never actually decoheres; what with being everything and all, it has no environment and there is therefore nowhere for information to escape to.
I'm not sure what "fundamental necessity" is supposed to mean or what it might mean to ascribe it to causation.
I'm saying that the most fundamental feature of any sound theory of how stuff actually works must rest squarely on the observation that something is happening, and that saying anything more precise than that about what is happening (or has, or will) requires making distinctions i.e. a conceptual carving-up of what's happening into distinguishable parts.
I'm saying that we have all been performing just such a carving-up since before we learned to speak. I think the fundamental feature of every person's personal physics is the distinction between "me" and "everything else", and that that's a distinction made so early that its influence over the rest of our understanding is inescapable unless we choose not to operate our brains in Safe Mode.
In particular, it seems to me that the idea of the universe as being composed of conceptually separable bodies rests squarely on the initial conceptual separation we've all made between our own bodies - that over which we can evidently exert control - and everything else, which we can't.
It also seems to me that the fundamental division of experience into past (that which we remember) and future (that which can surprise us) would have to happen very very early on as well.
Are you denying that causation (when it obtains) is a necessary connection between events?
Causation, loosely speaking, is the general principle that if this-kind-of-thing happens, that-kind-of-thing will surely follow. Before needing causation in order to give oneself an account of the world, one must invent not only distinctions but classification: the knowledge that this thing is one of those things. So it seems to me that causation is a principle that connects our internal models of events.
I don't believe that causation connects things in and of themselves in such a way as to rule out the possibility of events that have no cause even in principle, because I'm not convinced that the distinctions and classifications we need to make in order to identify events are in fact anything more than conveniences and conventions.
Or that causation is necessary for doing science?
Causation is undoubtedly a very useful principle, and works extremely well for acquiring and communicating understanding of a vast array of observed behaviour. But what I'm suggesting is that it's not the only useful conceptual tool applicable to understanding the correlations we observe in and around us, and that the dissatisfaction people experience on finding out that one of our best theoretical frameworks provides "mere" probability predictions rather than 100% repeatable if-this-then-that is actually completely unwarranted.
Because when it comes right down to it, the predictions made by every physical theory are probabilistic. There is no way to guarantee with 100% certainty that the conditions upon which you base the non-probabilistic predictions provided by a classical theory actually apply to any given piece of reality whose behaviour you are attempting to predict.
The beauty of quantum theory, to me, is in the way it wraps up that very uncertainty, incorporates it inside the theory itself, and turns it into probabilities good enough to do engineering with. It's an astounding intellectual achievement, and I strongly doubt I'll see it bettered before I die.
posted by flabdablet at 5:43 PM on July 10, 2017 [3 favorites]
I'm not convinced that the distinctions and classifications we need to make in order to identify events are in fact anything more than conveniences and conventions.
If we need them, then they're rather more than conveniences and conventions. Does a plant identify events and apply causal principles when it regulates its life cycle by seasonal cues? If it didn't do so, would it survive? Our current understanding of photosynthesis includes a sense of it as a supremely efficient quantum machine for manipulating energy transduced from photons. The plant doesn't need to model that, fair enough, but it's us doing the modelling. Do we need that model? Well, it gives us options to enhance the probability of our continued survival, which is apparently a perfectly valid need for a species.
At this point, I think the issue becomes one of the nature of conscious thought rather than physics. It literally doesn't make sense to consider an undifferentiated universe: true, the universe outside our head presumably doesn't 'think of itself' as causally-connected events intertwining discrete entities: it is what it is. But we cannot perceive or make any use of the universe in that form. Is our differentiated, categorising, predictive causal model of aspects of that universe, in some way 'true'? Lots of ink has been spilled on that, but for empiricism to work (and it bloody well does) then we're hoiking something of the actual nature of the universe out of it.
I think that as long as you're open to evidence that could potentially completely rewire your assumptions - as physics has been since Maxwell - going with what works is fine. 'Causality' may not be The Underlying Truth Of It All, but when you run your assumptions through the mill and out pops the predictions that match the CMB that was set in train 13 billion years ago, you really have to assign the idea a fairly high degree of probability that it's cooking with gas.
posted by Devonian at 6:20 PM on July 10, 2017 [2 favorites]
If we need them, then they're rather more than conveniences and conventions. Does a plant identify events and apply causal principles when it regulates its life cycle by seasonal cues? If it didn't do so, would it survive? Our current understanding of photosynthesis includes a sense of it as a supremely efficient quantum machine for manipulating energy transduced from photons. The plant doesn't need to model that, fair enough, but it's us doing the modelling. Do we need that model? Well, it gives us options to enhance the probability of our continued survival, which is apparently a perfectly valid need for a species.
At this point, I think the issue becomes one of the nature of conscious thought rather than physics. It literally doesn't make sense to consider an undifferentiated universe: true, the universe outside our head presumably doesn't 'think of itself' as causally-connected events intertwining discrete entities: it is what it is. But we cannot perceive or make any use of the universe in that form. Is our differentiated, categorising, predictive causal model of aspects of that universe, in some way 'true'? Lots of ink has been spilled on that, but for empiricism to work (and it bloody well does) then we're hoiking something of the actual nature of the universe out of it.
I think that as long as you're open to evidence that could potentially completely rewire your assumptions - as physics has been since Maxwell - going with what works is fine. 'Causality' may not be The Underlying Truth Of It All, but when you run your assumptions through the mill and out pops the predictions that match the CMB that was set in train 13 billion years ago, you really have to assign the idea a fairly high degree of probability that it's cooking with gas.
posted by Devonian at 6:20 PM on July 10, 2017 [2 favorites]
flabdablet: yes yes yes yes yes!
I think the fundamental feature of every person's personal physics is the distinction between "me" and "everything else"
I really have to recommend a book: Meeting the Universe Halfway by Karen Barad (Duke University Press, 2007) -- she is a lattice gauge theory PhD turned philosophy professor. The book attempts to communicate to philosophy types the real nature of the interaction between observer and observed in quantum physics, focusing on the problem of our usual "cut" between "me" and "everything else". Way more interesting than Penrose. ;)
posted by heatherlogan at 6:42 PM on July 10, 2017 [7 favorites]
I think the fundamental feature of every person's personal physics is the distinction between "me" and "everything else"
I really have to recommend a book: Meeting the Universe Halfway by Karen Barad (Duke University Press, 2007) -- she is a lattice gauge theory PhD turned philosophy professor. The book attempts to communicate to philosophy types the real nature of the interaction between observer and observed in quantum physics, focusing on the problem of our usual "cut" between "me" and "everything else". Way more interesting than Penrose. ;)
posted by heatherlogan at 6:42 PM on July 10, 2017 [7 favorites]
To expand heatherlogan's example a bit more, not only does making a sound shorter make the pitch less clear and the timing more clear, but making the pitch more clear means smoothing the start and end which makes the timing less clear. This comes up extensively in frequency based DSP, where algorithms with limited bit rate or CPU time available trade off time clarity (thing happens at just the right time) vs. pitch clarity (thing has the right fundamental and overtones). Think, for example, the Scylla and Charybdis that are bleepy noises in the cymbals and smeared percussion sounds that lack crispness in low bitrate mp3s.
posted by idiopath at 6:47 PM on July 10, 2017 [4 favorites]
posted by idiopath at 6:47 PM on July 10, 2017 [4 favorites]
If we need them, then they're rather more than conveniences and conventions.
We need them in order to understand the world we exist within and communicate those understandings amongst ourselves. It seems to me that a hypothetical world exactly like ours apart from a complete lack of beings like us ought to be completely feasible and behave much the same way regardless of the fact that nobody inside it had any need for our conveniences and conventions.
Does a plant identify events and apply causal principles when it regulates its life cycle by seasonal cues?
Seems unlikely to me, because I'm currently unaware of the existence within plants of any structure with comparable modelling power to that of a human brain.
If it didn't do so, would it survive?
Yes. As evidence, I offer the fossil record demonstrating the existence of plants literally billions of years before we turned up and began describing them to each other.
Our current understanding of photosynthesis includes a sense of it as a supremely efficient quantum machine for manipulating energy transduced from photons. The plant doesn't need to model that, fair enough, but it's us doing the modelling. Do we need that model? Well, it gives us options to enhance the probability of our continued survival, which is apparently a perfectly valid need for a species.
Again, we're back to the observation that it's us who has a need for causality, not the plants we use it to understand.
At this point, I think the issue becomes one of the nature of conscious thought rather than physics.
One of the things I find that I personally need is a clear understanding of the relationship between those two things. I'm currently quite comfortable with a physical model that makes consciousness a local property of certain physical structures bearing some degree of resemblance to human brains, and a conceptual model that considers the whole of physics to be part of what human consciousness does.
It literally doesn't make sense to consider an undifferentiated universe
That depends entirely what you mean by "make sense" and "consider". I have personally found that it makes a great deal of sense to spend at least some time on serious and increasingly successful attempts to experience the universe as undifferentiated.
Part of that has involved learning to let go of the naive expectation that such a process would necessarily end up conferring knowledge about regions of the physical universe that are inaccessible via the more customary states of mind.
There's an almost Heisenbergian relationship between the ability to experience the universe as undifferentiated and the ability to communicate the usefulness of that experience to anybody else :-)
Is our differentiated, categorising, predictive causal model of aspects of that universe, in some way 'true'?
That depends entirely what you mean by "true". If it means that the models we build in order to do engineering with actually work, then yes. If it means that the models are something other than models, then no.
for empiricism to work (and it bloody well does) then we're hoiking something of the actual nature of the universe out of it
Quite so. And the actual thing we're hoiking out is a set of bloody good models.
I think that as long as you're open to evidence that could potentially completely rewire your assumptions - as physics has been since Maxwell - going with what works is fine.
I think that going with what works is fine regardless of your attitude toward new evidence, because I think each of us has different criteria for "works" and I think attitude toward new evidence has to be part of that.
Seems to me that the real trick is coming up with a halfway decent cost/benefit analysis for any given collection of assumptions. I've tried throwing them all out; ended up tied down to a bed in psych ward, so that doesn't seem to be a good idea. "Everything I know might be wrong, but given that I'm still alive right now and reasonably happy most of the time, most of it probably isn't too far off" works pretty well as a bedrock assumption.
'Causality' may not be The Underlying Truth Of It All
...and here we are at heated agreement :-)
posted by flabdablet at 6:56 PM on July 10, 2017 [1 favorite]
We need them in order to understand the world we exist within and communicate those understandings amongst ourselves. It seems to me that a hypothetical world exactly like ours apart from a complete lack of beings like us ought to be completely feasible and behave much the same way regardless of the fact that nobody inside it had any need for our conveniences and conventions.
Does a plant identify events and apply causal principles when it regulates its life cycle by seasonal cues?
Seems unlikely to me, because I'm currently unaware of the existence within plants of any structure with comparable modelling power to that of a human brain.
If it didn't do so, would it survive?
Yes. As evidence, I offer the fossil record demonstrating the existence of plants literally billions of years before we turned up and began describing them to each other.
Our current understanding of photosynthesis includes a sense of it as a supremely efficient quantum machine for manipulating energy transduced from photons. The plant doesn't need to model that, fair enough, but it's us doing the modelling. Do we need that model? Well, it gives us options to enhance the probability of our continued survival, which is apparently a perfectly valid need for a species.
Again, we're back to the observation that it's us who has a need for causality, not the plants we use it to understand.
At this point, I think the issue becomes one of the nature of conscious thought rather than physics.
One of the things I find that I personally need is a clear understanding of the relationship between those two things. I'm currently quite comfortable with a physical model that makes consciousness a local property of certain physical structures bearing some degree of resemblance to human brains, and a conceptual model that considers the whole of physics to be part of what human consciousness does.
It literally doesn't make sense to consider an undifferentiated universe
That depends entirely what you mean by "make sense" and "consider". I have personally found that it makes a great deal of sense to spend at least some time on serious and increasingly successful attempts to experience the universe as undifferentiated.
Part of that has involved learning to let go of the naive expectation that such a process would necessarily end up conferring knowledge about regions of the physical universe that are inaccessible via the more customary states of mind.
There's an almost Heisenbergian relationship between the ability to experience the universe as undifferentiated and the ability to communicate the usefulness of that experience to anybody else :-)
Is our differentiated, categorising, predictive causal model of aspects of that universe, in some way 'true'?
That depends entirely what you mean by "true". If it means that the models we build in order to do engineering with actually work, then yes. If it means that the models are something other than models, then no.
for empiricism to work (and it bloody well does) then we're hoiking something of the actual nature of the universe out of it
Quite so. And the actual thing we're hoiking out is a set of bloody good models.
I think that as long as you're open to evidence that could potentially completely rewire your assumptions - as physics has been since Maxwell - going with what works is fine.
I think that going with what works is fine regardless of your attitude toward new evidence, because I think each of us has different criteria for "works" and I think attitude toward new evidence has to be part of that.
Seems to me that the real trick is coming up with a halfway decent cost/benefit analysis for any given collection of assumptions. I've tried throwing them all out; ended up tied down to a bed in psych ward, so that doesn't seem to be a good idea. "Everything I know might be wrong, but given that I'm still alive right now and reasonably happy most of the time, most of it probably isn't too far off" works pretty well as a bedrock assumption.
'Causality' may not be The Underlying Truth Of It All
...and here we are at heated agreement :-)
posted by flabdablet at 6:56 PM on July 10, 2017 [1 favorite]
Meeting the Universe Halfway
Added to the reading list. Thanks!
posted by flabdablet at 6:57 PM on July 10, 2017 [1 favorite]
Added to the reading list. Thanks!
posted by flabdablet at 6:57 PM on July 10, 2017 [1 favorite]
I can just about cope with holes in semiconductors behaving like particles, because even at a classical level that makes sense. Phonons, OK; you can look at stuff like eddies in a river and sure, they clearly have existence and interactions that make more sense if you think of them as things, rather than not-things, that follow similar patterns to other actual things. Electrons splitting into spinons, orbitons and for all I know, Mysterons - why not, if they fancy it. All just probabilistic perturbations of patterns in fields.
But you can entangle the not-things?
To somebody who has the traditionally "real" particles like electrons and photons filed in the same conceptual drawer as eddies and phonons, the fact that a successful theory for predicting the behaviour of the former is also to some degree applicable to the latter is entirely unsurprising.
At this point, I tend to start asking myself 'does the stuff produce the rules, or the rules produce the stuff', and considering a career change to acid-taster for the Dali Lama.
Recommended more on a hobby basis than as a career, but whatever works for you.
Also curious to know whether you end up agreeing that the stuff produces us, we produce rules to match the stuff, and that the only really weird part is the knots we seem to need to tie ourselves in to account for the observation that those rules are so often fit for purpose.
posted by flabdablet at 7:13 PM on July 10, 2017
But you can entangle the not-things?
To somebody who has the traditionally "real" particles like electrons and photons filed in the same conceptual drawer as eddies and phonons, the fact that a successful theory for predicting the behaviour of the former is also to some degree applicable to the latter is entirely unsurprising.
At this point, I tend to start asking myself 'does the stuff produce the rules, or the rules produce the stuff', and considering a career change to acid-taster for the Dali Lama.
Recommended more on a hobby basis than as a career, but whatever works for you.
Also curious to know whether you end up agreeing that the stuff produces us, we produce rules to match the stuff, and that the only really weird part is the knots we seem to need to tie ourselves in to account for the observation that those rules are so often fit for purpose.
posted by flabdablet at 7:13 PM on July 10, 2017
The book attempts to communicate to philosophy types the real nature of the interaction between observer and observed in quantum physics, focusing on the problem of our usual "cut" between "me" and "everything else".
Does it have useful insights to offer on the other fundamental conceptual cut, the one between past and future? That's the one that keeps me awake at nights.
If anybody could point me to a precise duration for now - in nanoseconds, please, with error bars - I'd appreciate it.
posted by flabdablet at 7:20 PM on July 10, 2017
Does it have useful insights to offer on the other fundamental conceptual cut, the one between past and future? That's the one that keeps me awake at nights.
If anybody could point me to a precise duration for now - in nanoseconds, please, with error bars - I'd appreciate it.
posted by flabdablet at 7:20 PM on July 10, 2017
Now, then: 5.39116(13)x10-35 ns
You're welcome.
posted by Devonian at 7:45 PM on July 10, 2017 [3 favorites]
You're welcome.
posted by Devonian at 7:45 PM on July 10, 2017 [3 favorites]
As for the previous discussion - I think we're probably having a non-discussion. I would only add that absolutely everything we do in our conscious mind is of necessity only a model (it is a silly place), so we can't import anything else; the question is how closely that model reflects the unknowable actuality - and this of necessity again includes any contemplation of the undifferentiated universe.
Such contemplation has many uses. Physics isn't a primary beneficiary, but I would argue that it can offer useful insights even here, by allowing easier perceptions of new patterns and dislodging, or at least recontextualising, old ones.
posted by Devonian at 7:54 PM on July 10, 2017 [2 favorites]
Such contemplation has many uses. Physics isn't a primary beneficiary, but I would argue that it can offer useful insights even here, by allowing easier perceptions of new patterns and dislodging, or at least recontextualising, old ones.
posted by Devonian at 7:54 PM on July 10, 2017 [2 favorites]
Does it have useful insights to offer on the other fundamental conceptual cut, the one between past and future?
A little. The model used is that of intra-actions (a neologism meant to emphasize that the interaction is not between two entities that are separate in the classical sense, but a big mash), which inherently take some time to happen.
In quantum mechanics language, you choose an initial state and a final state that you're interested in, stick the time translation operator in between, and there's your probability amplitude. In the limit that the time difference between initial and final goes to zero, all you've got is the projection of the initial state onto the final (the wavefunction overlap) with nothing happening in between. So "now" seems to have meaning, but it's too short a time for anything to actually happen.
posted by heatherlogan at 6:07 AM on July 11, 2017 [2 favorites]
A little. The model used is that of intra-actions (a neologism meant to emphasize that the interaction is not between two entities that are separate in the classical sense, but a big mash), which inherently take some time to happen.
In quantum mechanics language, you choose an initial state and a final state that you're interested in, stick the time translation operator in between, and there's your probability amplitude. In the limit that the time difference between initial and final goes to zero, all you've got is the projection of the initial state onto the final (the wavefunction overlap) with nothing happening in between. So "now" seems to have meaning, but it's too short a time for anything to actually happen.
posted by heatherlogan at 6:07 AM on July 11, 2017 [2 favorites]
By the way, has anyone read the article(s) yet? Besides the sciencealert piece, I mean. (I'm going to try to...)
posted by heatherlogan at 6:08 AM on July 11, 2017
posted by heatherlogan at 6:08 AM on July 11, 2017
One way to think about it is to make the analogy to sound waves.
that was very interesting, and news to me!
posted by thelonius at 12:11 AM on July 12, 2017
that was very interesting, and news to me!
posted by thelonius at 12:11 AM on July 12, 2017
The EPR postulate ingredients that have to be given up are either locality or "reality" ("reality" meaning that each particle is in some definite state, i.e., no superpositions).
heatherlogan: this is not true for the version of EPR that Einstein favored. (This can be found in his letters to Schrödinger in the wake of the EPR paper's publication in 1935, the key bits of which are collected and translated into English in Arthur Fine's The Shaky Game. The EPR paper itself is not a great place to start — Podolsky wrote it, and Einstein wrote to Schrodinger that "it did not come out as well as I had originally wanted; rather the essential thing was, so to speak, smothered by the [mathematics].")
The EPR argument is an argument from locality to determinism and the incompleteness of quantum mechanics. It argues that because these long-distance correlations (i.e. entanglement) exist, the only local way to account for them is to say that the quantum state isn't the whole story, that the entangled particles have individual well-determined states before they're measured despite the fact that the quantum wavefunction doesn't say what those states are. They don't postulate that the individual particles are in definite states all along — they infer it from locality and entanglement. Hence EPR's conclusion that, if locality is correct, quantum physics must be incomplete.
But, in the immortal words of LeVar Burton, you don't have to take my word for it. Just ask John Bell:
"It is important to note that to the limited degree to which determinism plays a role in the EPR argument, it is not assumed but inferred. What is held sacred is the principle of 'local causality' — or 'no action at a distance'....It is remarkably difficult to get this point across, that determinism is not a presupposition of the analysis." (Speakable and Unspeakable in Quantum Mechanics, p. 143, emphasis his.) Bell also pointed out that his own celebrated theorem was misunderstood in much the same way: his theorem starts, as EPR did, from an assumption of locality, and proceeds from there to infer deterministic 'hidden variables' that underlie the entangled quantum state. Yet it's often said that his theorem starts from the assumption of determinism, which is simply not true.
I have absolutely no idea what causal mechanisms could possibly operate inside an entangled pair in order to "make" the measurements come out the way they do, and have seen no evidence that anybody else does either...
flabdablet: There are certainly causal mechanisms that could be at work within an entangled pair that would lead to the measurements we see. Non-local hidden variable theories, like pilot-wave theory (aka de Broglie-Bohm theory, or Bohmian mechanics) handles entangled pairs (and the rest of bog-standard quantum mechanics) very well indeed. Same goes for retrocausal theories like the ones being discussed in the article at hand. As heatherlogan points out, the Schrödinger equation itself is a causal mechanism, and an adherent of the many-worlds interpretation would disagree with you that it doesn't predict specific outcomes, since (on that view) all the outcomes predicted by the Schrödinger equation actually come to pass. And so on.
I think the fundamental feature of every person's personal physics is the distinction between "me" and "everything else"...
flabdablet: So are you proposing that quantum physics requires a boundary in the world beyond which it doesn't function? If so, in what way is it a fundamental theory? And how do you account for the fact that you and I always get the same answers when we're doing quantum physics and looking at the outcomes of quantum physics experiments? How does this not collapse into solipsism?
More generally, the idea that there's a boundary beyond which quantum physics doesn't work is a really weird one, especially since it's clearly not an idea that's forced on us by the theory itself. The existence of non-Copenhagen interpretations (like the ones I mention above) makes it clear that the idea of quantum theory as something other than a universal theory, or a theory that necessarily involves splitting the world into observer and observed, is a deliberate theoretical choice that's not driven by the physics itself. You want to make that choice, go ahead — but you'll need to justify it with something other than quantum theory.
posted by freelanceastro at 3:49 PM on July 12, 2017 [1 favorite]
heatherlogan: this is not true for the version of EPR that Einstein favored. (This can be found in his letters to Schrödinger in the wake of the EPR paper's publication in 1935, the key bits of which are collected and translated into English in Arthur Fine's The Shaky Game. The EPR paper itself is not a great place to start — Podolsky wrote it, and Einstein wrote to Schrodinger that "it did not come out as well as I had originally wanted; rather the essential thing was, so to speak, smothered by the [mathematics].")
The EPR argument is an argument from locality to determinism and the incompleteness of quantum mechanics. It argues that because these long-distance correlations (i.e. entanglement) exist, the only local way to account for them is to say that the quantum state isn't the whole story, that the entangled particles have individual well-determined states before they're measured despite the fact that the quantum wavefunction doesn't say what those states are. They don't postulate that the individual particles are in definite states all along — they infer it from locality and entanglement. Hence EPR's conclusion that, if locality is correct, quantum physics must be incomplete.
But, in the immortal words of LeVar Burton, you don't have to take my word for it. Just ask John Bell:
"It is important to note that to the limited degree to which determinism plays a role in the EPR argument, it is not assumed but inferred. What is held sacred is the principle of 'local causality' — or 'no action at a distance'....It is remarkably difficult to get this point across, that determinism is not a presupposition of the analysis." (Speakable and Unspeakable in Quantum Mechanics, p. 143, emphasis his.) Bell also pointed out that his own celebrated theorem was misunderstood in much the same way: his theorem starts, as EPR did, from an assumption of locality, and proceeds from there to infer deterministic 'hidden variables' that underlie the entangled quantum state. Yet it's often said that his theorem starts from the assumption of determinism, which is simply not true.
I have absolutely no idea what causal mechanisms could possibly operate inside an entangled pair in order to "make" the measurements come out the way they do, and have seen no evidence that anybody else does either...
flabdablet: There are certainly causal mechanisms that could be at work within an entangled pair that would lead to the measurements we see. Non-local hidden variable theories, like pilot-wave theory (aka de Broglie-Bohm theory, or Bohmian mechanics) handles entangled pairs (and the rest of bog-standard quantum mechanics) very well indeed. Same goes for retrocausal theories like the ones being discussed in the article at hand. As heatherlogan points out, the Schrödinger equation itself is a causal mechanism, and an adherent of the many-worlds interpretation would disagree with you that it doesn't predict specific outcomes, since (on that view) all the outcomes predicted by the Schrödinger equation actually come to pass. And so on.
I think the fundamental feature of every person's personal physics is the distinction between "me" and "everything else"...
flabdablet: So are you proposing that quantum physics requires a boundary in the world beyond which it doesn't function? If so, in what way is it a fundamental theory? And how do you account for the fact that you and I always get the same answers when we're doing quantum physics and looking at the outcomes of quantum physics experiments? How does this not collapse into solipsism?
More generally, the idea that there's a boundary beyond which quantum physics doesn't work is a really weird one, especially since it's clearly not an idea that's forced on us by the theory itself. The existence of non-Copenhagen interpretations (like the ones I mention above) makes it clear that the idea of quantum theory as something other than a universal theory, or a theory that necessarily involves splitting the world into observer and observed, is a deliberate theoretical choice that's not driven by the physics itself. You want to make that choice, go ahead — but you'll need to justify it with something other than quantum theory.
posted by freelanceastro at 3:49 PM on July 12, 2017 [1 favorite]
There are certainly causal mechanisms that could be at work within an entangled pair that would lead to the measurements we see. Non-local hidden variable theories, like pilot-wave theory (aka de Broglie-Bohm theory, or Bohmian mechanics) handles entangled pairs (and the rest of bog-standard quantum mechanics) very well indeed.
Sure, but if the point of looking for internal causality is to preserve locality, non-local hidden variables are possibly not what one would hope to end up relying on.
are you proposing that quantum physics requires a boundary in the world beyond which it doesn't function?
No, I'm proposing that the principle inherent in the world being in principle separable into parts at all is an intuition arising from the very first distinction any of us ever makes, which is the distinction between our own bodies (over which we observe we can exert control and within which we experience sensation) and everything else (over which we can't and don't).
I'm proposing that atoms-in-a-void models are mathematically convenient but intuitively misleading.
This view is derived from the observation that identifying any two separate things as separate during the instant in which they are interacting doesn't work even intuitively unless they have mathematical jump discontinuities at their boundaries, and that this failure to work is even worse for things considered pointlike.
how do you account for the fact that you and I always get the same answers when we're doing quantum physics and looking at the outcomes of quantum physics experiments?
The point of quantum physics is precisely that not only do you and I not always get the same answers, we don't even necessarily get repeatable answers ourselves. We can accurately calculate probabilities for getting any given answer, but that's as good as Nature lets our predictions get.
To me, that's a pretty good indication that Nature, while exhibiting all kinds of statistical regularities, doesn't ever exactly repeat itself. No two snowflakes and so forth.
It seems plausible to me that a nonlocal hidden variable theory might be better described as a nonlocal relevant variable theory. I suspect that nonlocality is quite sufficient on its own to make the relevant variables non-obvious because they pertain to stuff that's elsewhere, rather than being reflective of properties hidden in any sense inside the subsystem under analysis.
How does this not collapse into solipsism?
There is no solipsism involved in the act of experiencing the universe as an undifferentiated whole without distinctions between parts. Solipsism is a philosophical viewpoint, and the only way to gain access to such an experience is to temporarily shut down those parts of the brain responsible for performing philosophy.
Best one can do without learning the specific skills required to achieve that shutdown is simply decide not to care about distinctions for a while.
Even then, the resulting worldview is not solipsistic. Solipsism is the view that the entire world is contained within the mind of the thinker; this necessarily involves a distinction between that which contains and that which is contained. The point of not caring about distinctions is that once that has been done, nothing at all can be said about the world (in fact not even thought about the world if the aforementioned skills are up to par).
As I mentioned upthread, communicating the usefulness of this exercise to those disinclined to value it spontaneously on the grounds of pure curiosity is exceedingly difficult.
posted by flabdablet at 6:33 AM on July 13, 2017
Sure, but if the point of looking for internal causality is to preserve locality, non-local hidden variables are possibly not what one would hope to end up relying on.
are you proposing that quantum physics requires a boundary in the world beyond which it doesn't function?
No, I'm proposing that the principle inherent in the world being in principle separable into parts at all is an intuition arising from the very first distinction any of us ever makes, which is the distinction between our own bodies (over which we observe we can exert control and within which we experience sensation) and everything else (over which we can't and don't).
I'm proposing that atoms-in-a-void models are mathematically convenient but intuitively misleading.
This view is derived from the observation that identifying any two separate things as separate during the instant in which they are interacting doesn't work even intuitively unless they have mathematical jump discontinuities at their boundaries, and that this failure to work is even worse for things considered pointlike.
how do you account for the fact that you and I always get the same answers when we're doing quantum physics and looking at the outcomes of quantum physics experiments?
The point of quantum physics is precisely that not only do you and I not always get the same answers, we don't even necessarily get repeatable answers ourselves. We can accurately calculate probabilities for getting any given answer, but that's as good as Nature lets our predictions get.
To me, that's a pretty good indication that Nature, while exhibiting all kinds of statistical regularities, doesn't ever exactly repeat itself. No two snowflakes and so forth.
It seems plausible to me that a nonlocal hidden variable theory might be better described as a nonlocal relevant variable theory. I suspect that nonlocality is quite sufficient on its own to make the relevant variables non-obvious because they pertain to stuff that's elsewhere, rather than being reflective of properties hidden in any sense inside the subsystem under analysis.
How does this not collapse into solipsism?
There is no solipsism involved in the act of experiencing the universe as an undifferentiated whole without distinctions between parts. Solipsism is a philosophical viewpoint, and the only way to gain access to such an experience is to temporarily shut down those parts of the brain responsible for performing philosophy.
Best one can do without learning the specific skills required to achieve that shutdown is simply decide not to care about distinctions for a while.
Even then, the resulting worldview is not solipsistic. Solipsism is the view that the entire world is contained within the mind of the thinker; this necessarily involves a distinction between that which contains and that which is contained. The point of not caring about distinctions is that once that has been done, nothing at all can be said about the world (in fact not even thought about the world if the aforementioned skills are up to par).
As I mentioned upthread, communicating the usefulness of this exercise to those disinclined to value it spontaneously on the grounds of pure curiosity is exceedingly difficult.
posted by flabdablet at 6:33 AM on July 13, 2017
I read arXiv:1510.06712. I read the entire introduction of arXiv:1607.07871, and skimmed the rest because that was all I could tolerate. If this represents cutting-edge research in "Quantum Foundations," then I can only conclude that the entire field is a heap of garbage. All they're doing in these papers is making up new "just-so" stories about what a particle is "really" doing, the predictions of which are entirely identical to the predictions of ordinary quantum mechanics, and which seem to add no further intuitive or pedagogical value.
From the FPP: Many people have heard of entanglement, an idea from quantum physics where two particles are connected in such a way that affecting one affects the other instantly, no matter where in the universe it is located. The fundamental mistake here is that affecting one such particle does not actually instantaneously affect the other. In light of special relativity, that would violate causality, which quantum mechanics does not do. It would also violate locality, and quantum mechanics, properly understood, is a strictly local theory. Observe:
Everyone probably knows the story of Schroedinger's Cat: You seal a (imaginary!) cat in a box with a radioactive source, a vial of poison, and a detection apparatus rigged to smash the vial if it detects a single radioactive decay. Now wait an appropriate amount of time such that the probability of a single detection event is 50%. At that moment, we say, the cat is in a mysterious superimposed state of being alive and dead, each with 50% probability. It is suspended in this weird quantum superposition until the experimenter opens the box, observing the cat and collapsing the wavefunction. This is the usual Copenhagen interpretation, also known as "lies told to children," because it leads to all kinds of pointless supposition about how the collapse can happen when it is not described by the Schroedinger equation and what the role of the conscious observer in causing the collapse could be, and what constitutes "conscious enough" (can a cat's observations collapse wavefunctions?).
In order to point out the problem with this simplistic interpretation, someone (maybe Wigner) came up with the story of Wigner's Friend: You have set up the Schroedinger's Cat experiment in the usual way, but now Wigner sends his friend into the lab to open the box and observe the cat. Upon doing so, Wigner's friend is herself now in a mysterious superimposed state of having observed a live cat and having observed a dead cat, each with 50% probability, and correlated with the appropriate parts of the wavefunction of the cat itself. Now Wigner's friend comes out of the lab and meets another friend of Wigner, and tells him what she observed. The second friend is now also in a mysterious superimposed state of having heard that the cat is alive and having heard that the cat is dead, again correlated with the quantum state of Wigner's first friend and the state of the cat itself. Et cetera.
Now apply this to two correlated particles. Imagine that Wigner causes to be produced two particles, whose wavefunctions are correlated (i.e., entangled) with each other -- for example, two photons in a state of zero total angular momentum. The first friend measures the polarization of one photon. Her wavefunction is now correlated with that photon's, which is in turn correlated with the other photon's. Around the same time, the second friend measures the wavefunction of the second photon. His wavefunction is now correlated with that of the second photon, and through it, with the first photon's and with the first friend's. When the two friends get together to discuss their results, of course they are going to agree, because their observations are appropriately correlated with each other. There is no such thing as collapse of the wavefunction, and all interactions are strictly local and causal.
Next we have to ask why Wigner's friend perceives herself as having observed only one particular outcome, when she is actually in a superposition of having observed different outcomes. This probably has to do with the nature of perception as looking back at one's own internal state, whereby the "looking back" gets correlated with the "having observed" state. And finally, we have to ask why the states we observed are those particular orthogonal projections. Apparently there is active debate about this -- see this very good Physics Today article by Wojciech Zurek and a letter to the editor in response. (Personally I think it's going to turn out to have something to do with which states are energy eigenstates of the macroscopic system together with the resulting rapid oscillations of all other linear combinations.)
posted by heatherlogan at 3:55 PM on July 19, 2017 [3 favorites]
From the FPP: Many people have heard of entanglement, an idea from quantum physics where two particles are connected in such a way that affecting one affects the other instantly, no matter where in the universe it is located. The fundamental mistake here is that affecting one such particle does not actually instantaneously affect the other. In light of special relativity, that would violate causality, which quantum mechanics does not do. It would also violate locality, and quantum mechanics, properly understood, is a strictly local theory. Observe:
Everyone probably knows the story of Schroedinger's Cat: You seal a (imaginary!) cat in a box with a radioactive source, a vial of poison, and a detection apparatus rigged to smash the vial if it detects a single radioactive decay. Now wait an appropriate amount of time such that the probability of a single detection event is 50%. At that moment, we say, the cat is in a mysterious superimposed state of being alive and dead, each with 50% probability. It is suspended in this weird quantum superposition until the experimenter opens the box, observing the cat and collapsing the wavefunction. This is the usual Copenhagen interpretation, also known as "lies told to children," because it leads to all kinds of pointless supposition about how the collapse can happen when it is not described by the Schroedinger equation and what the role of the conscious observer in causing the collapse could be, and what constitutes "conscious enough" (can a cat's observations collapse wavefunctions?).
In order to point out the problem with this simplistic interpretation, someone (maybe Wigner) came up with the story of Wigner's Friend: You have set up the Schroedinger's Cat experiment in the usual way, but now Wigner sends his friend into the lab to open the box and observe the cat. Upon doing so, Wigner's friend is herself now in a mysterious superimposed state of having observed a live cat and having observed a dead cat, each with 50% probability, and correlated with the appropriate parts of the wavefunction of the cat itself. Now Wigner's friend comes out of the lab and meets another friend of Wigner, and tells him what she observed. The second friend is now also in a mysterious superimposed state of having heard that the cat is alive and having heard that the cat is dead, again correlated with the quantum state of Wigner's first friend and the state of the cat itself. Et cetera.
Now apply this to two correlated particles. Imagine that Wigner causes to be produced two particles, whose wavefunctions are correlated (i.e., entangled) with each other -- for example, two photons in a state of zero total angular momentum. The first friend measures the polarization of one photon. Her wavefunction is now correlated with that photon's, which is in turn correlated with the other photon's. Around the same time, the second friend measures the wavefunction of the second photon. His wavefunction is now correlated with that of the second photon, and through it, with the first photon's and with the first friend's. When the two friends get together to discuss their results, of course they are going to agree, because their observations are appropriately correlated with each other. There is no such thing as collapse of the wavefunction, and all interactions are strictly local and causal.
Next we have to ask why Wigner's friend perceives herself as having observed only one particular outcome, when she is actually in a superposition of having observed different outcomes. This probably has to do with the nature of perception as looking back at one's own internal state, whereby the "looking back" gets correlated with the "having observed" state. And finally, we have to ask why the states we observed are those particular orthogonal projections. Apparently there is active debate about this -- see this very good Physics Today article by Wojciech Zurek and a letter to the editor in response. (Personally I think it's going to turn out to have something to do with which states are energy eigenstates of the macroscopic system together with the resulting rapid oscillations of all other linear combinations.)
posted by heatherlogan at 3:55 PM on July 19, 2017 [3 favorites]
All they're doing in these papers is making up new "just-so" stories about what a particle is "really" doing
...which is a testament to just how firmly so many people are attached to thinking of theories about reality as if they were reality.
"Particle" is a slippery word, because we use it both for directly perceptible features of reality such as dust motes and for the purely theoretical entities intuited and abstracted from those motes. Asking what a dust mote is "really" doing makes sense to me in a way that asking what an electron is "really" doing just doesn't.
What an electron is "really" doing is whatever the theory containing electrons says it's doing, provided that the theory is both applicable to and actually being applied to predict the behaviour of some part of reality as opposed to being exercised hypothetically.
we have to ask why Wigner's friend perceives herself as having observed only one particular outcome, when she is actually in a superposition of having observed different outcomes.
To me, that would simply be because observation always trumps prediction, for non-deluded observers.
The superposition of outcomes is part of a probabilistic model that provides the best available description of the outcomes of an ensemble of repeated experiments. But the thing about being a conscious being (perhaps even being a particle with a worldline) is that none of us ever really gets to repeat anything ever.
Life doesn't come with do-overs, not really. Closest we can get is doing a second thing that's enough like the first thing that ignoring the contextual differences between them is both reasonable and appropriate, which is exactly what controlling experimental conditions (and, by extension, engineering) is all about.
So it seems to me that the first question that needs to be asked about that superposition is where it is, and it further seems to me that the answer to that question is always going to be "inside some particular experimenter's mind". Reality is always just going to be whatever it is, regardless of what we think about that; when we say that some part of reality is in a superposition of possible states, what we're actually saying is that we have no way even in principle to describe it more precisely than that.
Wigner's friend is in a superposition of having observed different outcomes according to the best description of reality available even in principle to Wigner. But it's not all about Wigner.
all kinds of pointless supposition about how the collapse can happen when it is not described by the Schroedinger equation and what the role of the conscious observer in causing the collapse could be
Seems to me that most of the mystery there just kind of evaporates once serious attention is paid to the role of the conscious observer in causing the Schroedinger equation.
posted by flabdablet at 8:29 PM on July 19, 2017
...which is a testament to just how firmly so many people are attached to thinking of theories about reality as if they were reality.
"Particle" is a slippery word, because we use it both for directly perceptible features of reality such as dust motes and for the purely theoretical entities intuited and abstracted from those motes. Asking what a dust mote is "really" doing makes sense to me in a way that asking what an electron is "really" doing just doesn't.
What an electron is "really" doing is whatever the theory containing electrons says it's doing, provided that the theory is both applicable to and actually being applied to predict the behaviour of some part of reality as opposed to being exercised hypothetically.
we have to ask why Wigner's friend perceives herself as having observed only one particular outcome, when she is actually in a superposition of having observed different outcomes.
To me, that would simply be because observation always trumps prediction, for non-deluded observers.
The superposition of outcomes is part of a probabilistic model that provides the best available description of the outcomes of an ensemble of repeated experiments. But the thing about being a conscious being (perhaps even being a particle with a worldline) is that none of us ever really gets to repeat anything ever.
Life doesn't come with do-overs, not really. Closest we can get is doing a second thing that's enough like the first thing that ignoring the contextual differences between them is both reasonable and appropriate, which is exactly what controlling experimental conditions (and, by extension, engineering) is all about.
So it seems to me that the first question that needs to be asked about that superposition is where it is, and it further seems to me that the answer to that question is always going to be "inside some particular experimenter's mind". Reality is always just going to be whatever it is, regardless of what we think about that; when we say that some part of reality is in a superposition of possible states, what we're actually saying is that we have no way even in principle to describe it more precisely than that.
Wigner's friend is in a superposition of having observed different outcomes according to the best description of reality available even in principle to Wigner. But it's not all about Wigner.
all kinds of pointless supposition about how the collapse can happen when it is not described by the Schroedinger equation and what the role of the conscious observer in causing the collapse could be
Seems to me that most of the mystery there just kind of evaporates once serious attention is paid to the role of the conscious observer in causing the Schroedinger equation.
posted by flabdablet at 8:29 PM on July 19, 2017
It occurred to me last night that the answer to the question, "Is an electron a particle or a wave?" is "Mu!"
posted by heatherlogan at 3:45 AM on July 20, 2017 [2 favorites]
posted by heatherlogan at 3:45 AM on July 20, 2017 [2 favorites]
(Not to be confused with the muon, of course.)
posted by heatherlogan at 3:48 AM on July 20, 2017
posted by heatherlogan at 3:48 AM on July 20, 2017
« Older To save enough for a copy of Panzer Dragoon Saga | Meat, not too fresh. Newer »
This thread has been archived and is closed to new comments
posted by ZeusHumms at 6:32 AM on July 10, 2017 [20 favorites]