3 ms·

"local" seems like a fairly large assumption for something hypothesized by at least some to work through quantum mechanics, of which a central tenant is non-loc

by 3iak 11y ago

"local" seems like a fairly large assumption for something hypothesized by at least some to work through quantum mechanics, of which a central tenant is non-locality. In fact, given the consistency of qm experimental results over the past 50 or so years, it seems entirely plausible to think that it would be a field of study which could upend classical physics. All that being said I agree extreme skepticism is healthy here, though arrogance or dismissal doesn't seem quite appropriate as these results continue to be reproduced.

Steuard 11y ago
I may have been using a technical term too glibly there. Local conservation of momentum is a fundamental property of any translation-invariant model in the framework of quantum field theory. All this means is that the (vector) momentum flowing into any point in space-time must equal the momentum flowing out plus any increase in "stored" momentum at that point (whatever that might mean). Quantum mechanics is a "local" theory in the formal sense, at least in any halfway mainstream formulation (from Copenhagen to many-worlds). Even the "spooky action at a distance" that folks talk about in reference to quantum phenomena is never capable of transmitting information outside of the "cone of causality" in space-time (which is the "light cone" in most relativistic theories: information can't travel faster than light).
tjradcliffe 11y ago
Except the non-locality of QM is irrelevant to a device that is apparently a closed cavity. Simply because the same word is used does not mean the underlying concept is sufficiently similar in both contexts to be relevant. Local momentum conservation is foundational to all of physics: it is embedded in Newton's laws, and you can't get much more foundational than that. As such, if anything violates it the degree of "extraordinary" would be beyond anything ever seen. Far beyond superluminal nuetrinos. So the previous posters are correct: momentum conservation, which is generally used to mean "local momentum conservation", is as deeply foundational to physics as Newton's laws, and the kind of observable-only-by-statistical-inference non-locality seen in QM is not known to have any impact on it. As such, it would be earth-shaking at the deepest possible levels of physics if this result were correct. That does not mean it is wrong--although that is certainly my bet and that of virtually every other working physcist--but it's important not to understate what a big deal it would be, because understanding that opens up alternative ideas regarding how to test it. This is neither arrogance nor dismissal, but an understanding that anything that violates local momentum conservation, via QM or anything else, and also lets the rest of physcis sail blithely on, manifesting no measureable effect of any kind on any of the millions of diverse and precise experiments that have been performed, is astonishingly implausible. Could it happen? Sure. But it's really easy for laypeople to radically underestimate the implausibility of such claims, as was done all the time in the cold fusion community. Nor does reproduction of the effect say anything very much about either the underlying theory (such as it is) or the actual cause. Lots of things can create forces of micronewtons, and given how utterly implausible the claim is, most physicists are starting with a prior barely above epsilon. As such, the amount of evidence required to update our belief to something close to unity is going to be huge. Laypeople, who start with an unreasonably large prior because they are ignorant of how fine-tuned such novel physics would have to be to never show up anywhere else, will require less evidence.
gus_massa 11y ago
Quantum Mechanics is weird but it's not magic. One of the most unintuitive effects is entanglement. Almost all the entanglements and weird things of Quantum Mechanics are local. The electrons in a molecule have weird antisymmetric properties that mix different configurations (the easy picture is http://en.wikipedia.org/wiki/Resonance_(chemistry) http://en.wikipedia.org/wiki/Resonance_(chemistry) the mathematics is more complicated). You can entangle a few particles and move them at long distance, but if the particles hit another particle, the entanglement goes away. (You can transfer the entanglement, in very specific setups, but it's difficult, usually you get decoherence.) I don't know a way to transfer a lot of momentum form one point to a very distant point using entanglement. For example, you can put a box of particles in a rocket and each particle in that box in entangled with a particle that is in a box in Earth. When you measure the momentum of the particle in the rocket, it "changes" the momentum in the particle in Earth, so you can think that some momentum was "transferred". (This is not technically correct, but is a good enough approximation.) The problem is that the momentum that you measure is in a random distribution, and half of the time you'll "transfer" momentum in one direction to make your rocket faster and the other half of the times the momentum will go in the other direction to make your rocket slower. It's impossible to transfer information using entanglement, and it's impossible to select the direction of the "transferred" momentum, in particular because that could be used to transfer information.

by 3iak 11y ago

Steuard 11y ago

I may have been using a technical term too glibly there. Local conservation of momentum is a fundamental property of any translation-invariant model in the framework of quantum field theory. All this means is that the (vector) momentum flowing into any point in space-time must equal the momentum flowing out plus any increase in "stored" momentum at that point (whatever that might mean). Quantum mechanics is a "local" theory in the formal sense, at least in any halfway mainstream formulation (from Copenhagen to many-worlds). Even the "spooky action at a distance" that folks talk about in reference to quantum phenomena is never capable of transmitting information outside of the "cone of causality" in space-time (which is the "light cone" in most relativistic theories: information can't travel faster than light).

tjradcliffe 11y ago

Except the non-locality of QM is irrelevant to a device that is apparently a closed cavity. Simply because the same word is used does not mean the underlying concept is sufficiently similar in both contexts to be relevant. Local momentum conservation is foundational to all of physics: it is embedded in Newton's laws, and you can't get much more foundational than that. As such, if anything violates it the degree of "extraordinary" would be beyond anything ever seen. Far beyond superluminal nuetrinos. So the previous posters are correct: momentum conservation, which is generally used to mean "local momentum conservation", is as deeply foundational to physics as Newton's laws, and the kind of observable-only-by-statistical-inference non-locality seen in QM is not known to have any impact on it. As such, it would be earth-shaking at the deepest possible levels of physics if this result were correct. That does not mean it is wrong--although that is certainly my bet and that of virtually every other working physcist--but it's important not to understate what a big deal it would be, because understanding that opens up alternative ideas regarding how to test it. This is neither arrogance nor dismissal, but an understanding that anything that violates local momentum conservation, via QM or anything else, and also lets the rest of physcis sail blithely on, manifesting no measureable effect of any kind on any of the millions of diverse and precise experiments that have been performed, is astonishingly implausible. Could it happen? Sure. But it's really easy for laypeople to radically underestimate the implausibility of such claims, as was done all the time in the cold fusion community. Nor does reproduction of the effect say anything very much about either the underlying theory (such as it is) or the actual cause. Lots of things can create forces of micronewtons, and given how utterly implausible the claim is, most physicists are starting with a prior barely above epsilon. As such, the amount of evidence required to update our belief to something close to unity is going to be huge. Laypeople, who start with an unreasonably large prior because they are ignorant of how fine-tuned such novel physics would have to be to never show up anywhere else, will require less evidence.

gus_massa 11y ago

Quantum Mechanics is weird but it's not magic. One of the most unintuitive effects is entanglement. Almost all the entanglements and weird things of Quantum Mechanics are local. The electrons in a molecule have weird antisymmetric properties that mix different configurations (the easy picture is http://en.wikipedia.org/wiki/Resonance_(chemistry) http://en.wikipedia.org/wiki/Resonance_(chemistry) the mathematics is more complicated). You can entangle a few particles and move them at long distance, but if the particles hit another particle, the entanglement goes away. (You can transfer the entanglement, in very specific setups, but it's difficult, usually you get decoherence.) I don't know a way to transfer a lot of momentum form one point to a very distant point using entanglement. For example, you can put a box of particles in a rocket and each particle in that box in entangled with a particle that is in a box in Earth. When you measure the momentum of the particle in the rocket, it "changes" the momentum in the particle in Earth, so you can think that some momentum was "transferred". (This is not technically correct, but is a good enough approximation.) The problem is that the momentum that you measure is in a random distribution, and half of the time you'll "transfer" momentum in one direction to make your rocket faster and the other half of the times the momentum will go in the other direction to make your rocket slower. It's impossible to transfer information using entanglement, and it's impossible to select the direction of the "transferred" momentum, in particular because that could be used to transfer information.