Join 3,497 readers in helping fund MetaFilter (Hide)


Experimental delayed-choice entanglement swapping.
April 25, 2012 12:47 AM   Subscribe

"Using four photons, we can actively delay the choice of measurement on two of the photons into the time-like future of the registration of the other two photons. This effectively projects the two already registered photons onto one of two mutually exclusive quantum states in which the photons are either entangled (quantum correlations) or separable (classical correlations). This can also be viewed as ‘quantum steering into the past’." (arXiv, Nature Physics, Ars Technica)
posted by jeffburdges (80 comments total) 9 users marked this as a favorite

 
Oh entanglement, you are bizarre.
posted by Chekhovian at 1:02 AM on April 25, 2012


Neat!

You only need to stretch Victor's decision a few milliseconds into the future and you can clean up day trading.

On the other hand as with most things out of the quantum end of science it will require a close analysis of the vocabulary to believe that they're saying what it appears their saying. The unfortunately named "Quantum Teleportation" has caused quite a few misunderstandings.
posted by Tell Me No Lies at 1:06 AM on April 25, 2012 [1 favorite]



their ----> they're

editing window ---> strangely appropriate request in this thread
posted by Tell Me No Lies at 1:08 AM on April 25, 2012 [9 favorites]


Could someone explain to me why this isn't just a feature of conditioning on a common effect? After all, for any causal structure of the form A --> V <-- B over variables A, B, and V, conditioning on V induces an association between A and B.
posted by Jonathan Livengood at 1:09 AM on April 25, 2012


Scientific Mumbo-Jumbo: even when it's technically accurate, it still sounds like something that would have been laughed out of a Star Trek script.
posted by ShutterBun at 2:00 AM on April 25, 2012 [1 favorite]


Okay, here's how I understand the experiment:

+---------+                 +--------------+
|photon   |                 |polarization  |
|source 1 |----photon 1---->|measurer    1 |
+---------+                 +--------------+
      \
       \                                             +-----------+
        --------------photon 2---------------------->| Random    |
                                                     | entangler |
        --------------photon 3---------------------->|           |
      /                                              +------------
     /
+---------+                 +--------------+
|photon   |                 |polarization  |
|source 2 |----photon 4---->|measurer 2    |
+---------+                 +--------------+

=(Time A) =========>>>>========= (Time B) =======>>>=== (Time C) =====>>===


a)  photon 1 and photon 2 are produced entangled with a random polarization
b)  photon 3 and photon 4 are produced entangled with a random polarization

IFF the the random entangler does *NOT* entangle photons 2 and 3 then
the polarization measurements of photons 1 & 4 are not correlated.  I.E.
you get results consistent with the photon sources producing random
polarization.

IFF the random entangler *DOES* entangle photons 2 and 3 then the 
polarization measurements of photons 1 & 4 are perfectly correlated.  
In addition they match the polarization of the now entangled photons 2 and 3.

The key here is that a random chance at (Time C) appears to determine
the polarizations measured at (Time B).

Is that of any help?
posted by Tell Me No Lies at 2:07 AM on April 25, 2012 [17 favorites]


Not really. The diagram made me thing of Maud, and th

Hi! Obiwanwasabi. Say, that's a swell diagra
posted by obiwanwasabi at 2:18 AM on April 25, 2012


It appears a "quantum random number generator" controls whether Victor entangles his two photons when measuring them, meaning yes the overall experimental setup is conditioning on his random number generator, but seemingly not his actual measurement results.

In essence, Alice and Bob's results are "conditioned on a common cause", but that cause occurs in their future. There is no "arrow of time" for the photons though because Schrödinger's equation is time-symmetric.
posted by jeffburdges at 2:22 AM on April 25, 2012 [2 favorites]


"There is a theory which states that if ever anybody discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarre and inexplicable. There is another theory which states that this has already happened."

-- Douglas Adams

I think we've found evidence for the second theory.
posted by DreamerFi at 2:43 AM on April 25, 2012 [1 favorite]


Using four photons, corresponding to each of the four suits in a deck of cards, we can actively shuffle any of the two of the photons into the time-like future where you can pick one or the other of two cards.

Fucking card tricks.
posted by twoleftfeet at 2:58 AM on April 25, 2012


@Tell Me No Lies,
can you please explain whether this means i can send information backward in time?

it seem like:

if at time C random entanglement is true then at B correlation = true

if at time C random entanglement is false then at B correlation = maybe

so if I set up a stupendous number of them in parallel then and compared them all

if at time C random entanglement is false then at B correlation = probably no.

surely there is a sneaky quantum twist eg. you cant actually find out what happened at b until after c?
posted by compound eye at 3:03 AM on April 25, 2012 [2 favorites]


I wonder what will happen if you use the correlation measurement to trigger the actual entanglement choice (that sets the correlation measurement in the past) to the inverse of what the measurement predicts.

If predicts is the right word. I suspect that this may be closer to the truth. Or may have been.

Photons are weird. Being a photon, where you aren't in time at all (there's no time for anything moving at c, right?), must give you a very... different view of the universe.
posted by Devonian at 3:39 AM on April 25, 2012


So, could one set up a bunch of these in a cascade to predict the future on a longer scale?
posted by mccarty.tim at 3:48 AM on April 25, 2012


Assuming you had a lot of space because light speed...
posted by mccarty.tim at 3:51 AM on April 25, 2012


"In our experiment, the primary events are the polarization measurements of photons 1 and 4 by Alice and Bob. They keep their data sets for future evaluation. Each of these data sets by itself and their correlations are completely random and show no structure whatsoever. The other two photons (photons 2 and 3) are delayed until after Alice and Bob’s measurements, and sent to Victor for measurement. His measurement then decides the context and determines the interpretation of Alice and Bob’s data."

This sounds like the usual delayed choice quantum eraser experiments, and indicates there's no communication back in time - you need the list of Victor's choices to see the correlations.

Causality is still safe. Or at least no more at risk than it was last year.
posted by edd at 3:53 AM on April 25, 2012 [1 favorite]


we can actively delay the choice of measurement on two of the photons into the time-like future

'Time-like future', eh?

I project that I will have been reading this article on some tomorrow-like day when I probably will have had more spare moment-like resources to be doing so, but I may have been going to have crossed that bridge when I have been about to have gotten to it, if not before.
 
posted by Herodios at 4:03 AM on April 25, 2012 [1 favorite]


edd: But I wonder if it actually is. Let's say Victor turns his entangler on or off based on a coin flip, turning it on if the coin is heads, off if it's tails. And let's say we use one of the various techniques to slow photons down, such that Victor flips the coin and makes a decision 5 seconds in the future. If Alice and Bob detect the same polarization, they bet that the coin comes up heads, otherwise tails.

If I understand the explanation correctly, this means that Alice and Bob should be able to predict the fall of a coin with much better than normal odds, no?

If you could get five seconds of warning on, say, stock price movements.....
posted by Malor at 4:13 AM on April 25, 2012


The non-causation-messing apparently-faster-than-light entanglement stuff... let's see if I can get this right.

Alice and Bob meet on a planet where there's a bag with two balls in, one black, one white. Alice takes one of the balls, but keeps it hidden and doesn't look at it. Bob just takes the bag. They get into their respective spaceships and travel away from each other.

After an hour, they stop. Bob looks in the bag and Alice looks at her ball. At that instant, Bob can accurately predict Alice's observation, despite the system never having been measured before and no signal passing between them or being able to pass between them in time.
posted by Devonian at 4:28 AM on April 25, 2012


No, I think the point is that Victor makes a decision about the nature of his measurement, and that with the result of that measurement you can unravel the correlations in Alice and Bob's measurements. So you can't make the bet until you know the result of Victor's measurement.

Victor might decide to entangle Alice and Bob's photons by making an entanglement measurement, but until they know whether Victor's measurement showed a |φ+>23 or a |φ->23 state they can't infer anything.

At least that's how it's reading to me.
posted by edd at 4:31 AM on April 25, 2012 [1 favorite]


How long are we going to be puzzled about the "time travel" aspects of entanglement before we are willing to accept the supposedly-complicated view of a branching multiverse?
posted by DU at 4:44 AM on April 25, 2012


See? Told you I could do it.
posted by Marisa Stole the Precious Thing at 4:55 AM on April 25, 2012 [4 favorites]


There is no way of knowing which subset of Alice and Bob's measurements are correlated until Victor tells you, in this experiment at least. So there is no way of transmitting information backwards.

But what if victor entangled every particle? Or every third? Or some other agreed upon pattern?
posted by empath at 5:19 AM on April 25, 2012


If it's agreed upon then Alice and Bob know in advance, and they're still not getting information transmitted except by the conventional channels they used to make that agreement.
posted by edd at 5:24 AM on April 25, 2012


Oh, derp.
posted by empath at 5:44 AM on April 25, 2012


Alice and Bob meet on a planet where there's a bag with two balls in, one black, one white.

It's a bit weirder than that. Imagine they each take a coin. If Alice flips heads, Bob will flip tails. Fine.

Now imagine that Alice's coin lands edge up. So will Bob's.
posted by empath at 5:50 AM on April 25, 2012


Well, but what if Victor only entangles when AAPL is going up, and doesn't entangle when it's going down? It sure looks to me that Alice and Bob could see that suddenly their photons were correlating strongly, and could then buy AAPL before it went up, or sell it before it went down.
posted by Malor at 6:20 AM on April 25, 2012


Following up.... they'd have to be sending, say, 20 photons a second, and then if they started seeing strong correlations, they'd issue buy orders, and then when the correlations stopped, they'd issue sells. It wouldn't be perfect, but if they just averaged the last twenty photon measurements, I think they'd have a nearly perfect AAPL foretelling device, wouldn't they?

Does Victor even need to make a measurement? Doesn't he just need to entangle or not entangle? And the experiment says 'Alice and Bob can't communicate', but do we know that?
posted by Malor at 6:22 AM on April 25, 2012


DU: it's an epistemology problem. If it is a Multiverse we can't measure or detect it, so nothing about it can be science. If we can do science about it (it is measurable and detectable), it isn't a multiverse. Multiverse is interesting for science-inspired philosophy, but tautalogically nothing about the presence or absence of a multiverse can be scientific.
posted by idiopath at 6:23 AM on April 25, 2012 [1 favorite]


If it is a Multiverse we can't measure or detect it, so nothing about it can be science. If we can do science about it (it is measurable and detectable), it isn't a multiverse.

Except in experiments like these, where we can detect the effect of the multiverse without detecting it directly.

It's like the story of finding Neptune. Was Neptune only science once it had been seen directly? Or can you study the perturbations and come to the most parsimonious conclusion that it must exist without ever seeing it?
posted by DU at 6:27 AM on April 25, 2012


A key part of the conclusion from the Nature paper:

"Whether these two particles are entangled or separable has been decided after they have been measured. If one viewed the quantum state as a real physical object, one could get the paradoxical situation that future actions seem to have an influence on past and already irrevocably recorded events. However, there is never a paradox if the quantum state is viewed as no more than a ‘catalogue of our knowledge’[2]. Then the state is a probability list for all possible measurement outcomes, the relative temporal order of the three observers’ events is irrelevant and no physical interactions whatsoever between these events, especially into the past, are necessary to explain the delayed-choice entanglement swapping."
posted by Fat Charlie the Archangel at 6:38 AM on April 25, 2012 [2 favorites]


@Tell Me No Lies,
can you please explain whether this means i can send information backward in time?


On the face of it yes. However, see my earlier note on the difficulties of getting clear communication out of people doing quantum research. There is this tendency to re-use commonly understood words in misleading ways.

On the other hand I don't think we have actual words for a lot of what they're seeing, so I understand the attempts at analogy.
posted by Tell Me No Lies at 6:42 AM on April 25, 2012


Except in experiments like these, where we can detect the effect of the multiverse without detecting it directly.

Except that experiments like these are explained equally well by the other interpretations of QM. Many Worlds (Multiverse) doesn't predict the out come of these experiments any better than the Copenhagen Interpretation, Pilot Wave theory, the Transactional interpretation, etc.
posted by Bort at 6:42 AM on April 25, 2012 [1 favorite]


I'm also confused by the experiment. It seems like from the description, Alice and Bob can actually measure their photons before Victor performs his choice-- as in observe them and write down their polarization, which seems like it should break any entanglement.
posted by justkevin at 6:45 AM on April 25, 2012


Parsimonious? Forget multiplying entities, a multiverse assumes a massive exponentiation of entities. Billions of copies of the entire observable world every second? And recursively each of those branching at the same rate? By comparison, luminous aether is small potatoes.
posted by idiopath at 6:46 AM on April 25, 2012


Malor: there basically aren't correlations unless you know the result of Victor's measurement. Victor could make an entanglement measurement that says Alice and Bob's should be correlated one way, but that measurement could be such that they should be anticorrelated.

It's a bit like Alice and Bob's measurements form one half of a one-time pad and Victor's measurements form the other, but Victor writes the message the combined halves give in the future of when Alice and Bob measured their half.

This analogy won't be entirely accurate, but I think it still helps.
posted by edd at 6:46 AM on April 25, 2012


Alice and Bob can actually measure their photons before Victor performs his choice-- as in observe them and write down their polarization, which seems like it should break any entanglement.

They aren't entangled when they measure them.
posted by empath at 6:47 AM on April 25, 2012


And measuring doesn't break the entanglement anyway, it only collapses the wave function so the other particle is determined.
posted by empath at 6:48 AM on April 25, 2012


Reading these comments on my phone so just wanted to share that the comment here getting mangled in the conversion is deliciously appropriate.
posted by RolandOfEld at 6:52 AM on April 25, 2012


I believe the hardest question here is : Victor cannot send macroscopic information into the past this way by using macroscopic decision making rather than quantum randomness. Why not?

I've honestly never encountered a particularly compelling case explaining why not, frequently the results represent a "catalogue of our knowledge". Yet, occasionally one wonders if it can be built.

I assume you cannot messages into the past this way, but maybe only because "the numbers just don't work out", not the no-communication theorem is actually correct.. or maybe a more subtle no-communication theorem has yet to be found.
posted by jeffburdges at 6:53 AM on April 25, 2012


This looks quite similar to the delayed choice quantum eraser experiment
posted by crayz at 6:54 AM on April 25, 2012


crazy: Yes it's very similar, and that analogy is drawn in the paper, with the addition:
using a four-partite entangled state, allows to a posteriori decide a two-particle characteristic and thus has qualitatively new features. Just as there is a wave-particle duality for single particles, there is an entanglement-separability duality for two particles.
It's that this version doesn't operate on the behaviour of one particle, but on a pair of particles.
posted by edd at 6:58 AM on April 25, 2012


Parsimonious? Forget multiplying entities, a multiverse assumes a massive exponentiation of entities. Billions of copies of the entire observable world every second? And recursively each of those branching at the same rate? By comparison, luminous aether is small potatoes.

The simplest explanation is not necessarily the one with the least "stuff". What lies beyond the edge of the observable universe? We have no idea, we can't see it. But it would be quite surprising if it were "nothing".

A better answer is that it continues to extend more or less homogeneously, possibly indefinitely, even though that adds a lot more "stuff" and quite possibly an infinite amount more stuff.

Why? Because the "nothing" answer would require a complicated explanation as to why the universe should stop right at the edge of where we happen to be able to view. "Many words" versus "many worlds" as Max Tegmark described it.
posted by justkevin at 6:59 AM on April 25, 2012 [1 favorite]


Parsimonious? Forget multiplying entities, a multiverse assumes a massive exponentiation of entities.

A multiverse contains exactly zero additional entities. Whereas "instantaneous communication" is not only an additional entity but a pretty hard one to swallow.
posted by DU at 7:04 AM on April 25, 2012


Yeah, just skimming through the paper. Their little analysis is pretty readable:
With our ideal realization of the delayed-choice entanglement swapping gedanken experiment, we have demonstrated a generalization of Wheeler’s “delayed-choice” tests, going from the wave-particle duality of a single particle to the entanglement-separability duality of two particles. Whether these two particles are entangled or separable has been decided after they have been measured. If one views the quantum state as a real physical object, one could get the seemingly paradoxical situation that future actions appear as having an influence on past and already irrevocably recorded events. However, there is never a paradox if the quantum state is viewed as to be no more than a “catalogue of our knowledge”. Then the state is a probability list for all possible measurement outcomes, the relative temporal order of the three observer’s events is irrelevant and no physical interactions whatsoever between these events, especially into the past, are necessary to explain the delayed-choice entanglement swapping.
The idea that "real physical objects" exist is one of our last, best myths
posted by crayz at 7:08 AM on April 25, 2012 [1 favorite]


Whereas "instantaneous communication" is not only an additional entity but a pretty hard one to swallow.

That's interesting, I have the exact opposite belief. I find Many Worlds hard to swallow, but have no problem with non-locality. Though I don't see any evidence pointing to one over the other.
posted by Bort at 7:20 AM on April 25, 2012


The idea that "real physical objects" exist is one of our last, best myths

I think you'd have to be more specific about what you mean by 'exists'. Temperature is real and measurable, even though it's just a statistical phenomenon, rather than some inherent property of individual particles.
posted by empath at 7:22 AM on April 25, 2012


"What lies beyond the edge of the observable universe? We have no idea, we can't see it. But it would be quite surprising if it were "nothing"."

A whole diverging copy of our universe, one which by definition none of us will ever see, is not a simple explanation. It assumes a whole category of event which no human being has experienced or can experience, and can lead to no testable predictions. It may be conceptually simpler, and in that regard may be good philosophy, but it is by definiton not science.
posted by idiopath at 7:25 AM on April 25, 2012


I think you'd have to be more specific about what you mean by 'exists'

I mean, any time we try to describe an object in the world, when we look more closely it in fact appears to be an ongoing process that can never be fully understood/predicted. We conceptualize a world of separate things persisting through time and interacting with each other, when it's more like inseparable, entangled processes whose general macroscopic stability tricks us into missing the underlying processes

People seem desperate to believe that this is just some bong-hit philosophy or tiny, mostly inconsequential quantum weirdness, because the concept of objects persisting through time is deeply embedded in our view of the world and ourselves. I think Zellinger (who was part of this experiment) among others is specifically trying to disprove that:
For me. the main purpose of doing experiments is to show people how strange quantum physics is. Most physicists are very naive; most still believe in real waves or particles.
posted by crayz at 7:40 AM on April 25, 2012 [2 favorites]


I will now post a comment in the past, within this very thread.
posted by Marisa Stole the Precious Thing at 7:41 AM on April 25, 2012 [6 favorites]


Well, now we have to rely on the mods to repair causality and the fabric of spacetime (as if they don't have enough on their plate).
posted by edd at 7:55 AM on April 25, 2012


Temperature is an excellent metaphor, actually. Reconceptualizing "an electron" to be literally nothing more than a predictable statistical phenomenon
posted by crayz at 8:07 AM on April 25, 2012


A whole diverging copy of our universe, one which by definition none of us will ever see, is not a simple explanation. It assumes a whole category of event which no human being has experienced or can experience, and can lead to no testable predictions. It may be conceptually simpler, and in that regard may be good philosophy, but it is by definiton not science.

We can certainly see the other universes. The double-slit experiment is one example. In the MWI interpretation, you are regarding the effect of photons in other universes in the interference pattern.

You can explain the effect you are seeing without other universes, but your explanation will invoke an "event which no human being has experienced or can experience, and can lead to no testable predictions."

The eternal inflation model of cosmology is unrelated to MWI, except that it also predicts unconnected universes. These "bubble universes" may have completely different physical laws from our own. These bubbles are will never be detected, but the theory that predicts them agrees with what we see in our universe.
posted by justkevin at 8:08 AM on April 25, 2012 [1 favorite]


I mean, any time we try to describe an object in the world, when we look more closely it in fact appears to be an ongoing process that can never be fully understood/predicted. We conceptualize a world of separate things persisting through time and interacting with each other, when it's more like inseparable, entangled processes whose general macroscopic stability tricks us into missing the underlying processes

That's reductionism, and it's not the only way of looking at the world, though it can often be a useful way of looking at the world.

Take the action of a molecule of water in a river -- would it make sense to ignore the fact that it's part of a river, when the existence of the whirlpool clearly has an impact on the motion of the molecule? And if it the river has a measurable effect, then how is it not real?
posted by empath at 8:14 AM on April 25, 2012 [1 favorite]


I find Many Worlds hard to swallow, but have no problem with non-locality. Though I don't see any evidence pointing to one over the other.

Well, there's Occam. As noted previously, "more mass" != "more entities". But "there's this thing called non-locality" is.
posted by DU at 8:20 AM on April 25, 2012


That's reductionism, and it's not the only way of looking at the world, though it can often be a useful way of looking at the world.

Reductionism in physical science is materialism, the idea that reality can be reduced to physical parts. Quantum experiments are a strong argument that materialism is simply not true. Epicycles were also a useful way of looking at the world
posted by crayz at 8:36 AM on April 25, 2012


Marisa Stole the Precious Thing: that was extremely well done.
posted by andreaazure at 8:46 AM on April 25, 2012 [1 favorite]


Quantum experiments are a strong argument that materialism is simply not true.

That's a highly controversial assertion.

If you're defining materialism as: "In philosophy, the theory of materialism holds that the only thing that exists is matter or energy; that all things are composed of material and all phenomena (including consciousness) are the result of material interactions." then I don't see how QM experiments indicate that materialism is invalid.
posted by Bort at 8:59 AM on April 25, 2012


Doesn't Shor's Algorithm rather suggest the multiverse interpretation is likely?

I spent a lot of time recently trying to come up with a description of a photon that I was happy with, especially when applied to radio frequency EM radiation (which is _always_ thought of as Maxwellian rather than quantum). Couldn't do it - I understand that we can describe it well in maths (which I appreciate, rather than comprehend), but there is no classical mental model that works. Particles, waves, packets of waves, probability matrix...

Still gnawing at it. There is a small but real danger of obsession here. Any pointers to particularly lucid discussions of the nature of photons welcome...
posted by Devonian at 9:17 AM on April 25, 2012


Doesn't Shor's Algorithm rather suggest the multiverse interpretation is likely?

No.
posted by Bort at 9:32 AM on April 25, 2012


The problem with all this is that everyone still has to hustle all these photons around. From the point of view of someone trying to build a superior interplanetary comms system, say, that's really not convenient.

If the photon sources, and the measurers, are on planet Earth, say, and the entangler is maybe on Titan, or Io, you still need for there to be a steady stream (actually two separate steady streams) of photons from the sources to the entanglers just to get started on attempting this.

I guess that's why they probably tried it on somebody's lab bench first.
posted by newdaddy at 9:34 AM on April 25, 2012


Ahh, there is a fairly simple explanation of why Victor cannot send macroscopic information backwards in time, namely his decision determines whether Alice and Bob's photons are uncorrelated vs. correlated or anti-correlated. Alice and Bob cannot distinguish between uncorrelated vs. correlated or anti-correlated without knowing Victor's results.

It remains unclear to me if you can deduce this from the No-communication theorem by projecting this complex system onto the simpler system appearing in the No-communication theorem.
posted by jeffburdges at 9:34 AM on April 25, 2012


"Alice and Bob cannot distinguish between uncorrelated vs. correlated or anti-correlated without knowing Victor's results. " - Is that true?

I'm thinking there should be some kind of transitive behavior with correlation here - if photon 1 is correlated with 2 (and in this experment, we know that it is), and 2 is correlated with photon 3, and 3 is correlated with 4 (again, this is given), then 1 is correlated with 4.

I don't think Alice and Bob need to chat with Victor at all - they can just compare their own measurements for photons 1 and 4.

It's entirely possible I'm misunderstanding this - I've not read the actual paper.
posted by newdaddy at 9:42 AM on April 25, 2012


newdaddy: the entanglement is swapped so the correlation between 1 and 3, and between 2 and 4 disappears and correlations between 1 and 4, and between 2 and 3 take their place.
posted by edd at 10:07 AM on April 25, 2012


(may have interchanged some photon labels there but I hope that's still clear)
posted by edd at 10:12 AM on April 25, 2012


Let's imagine that you have two strings of randomly generated bits. If you compare them bit to bit, sometimes you will have two ones, sometimes you will have two zeros and sometimes you will have a one and a zero.

From any subset of randomly generated bits you can choose many subsets which appear to be correlated or anti correlated. There is no way to distinguish those randomly correlated subsets from the actual subset that was correlated because of entanglement without the list of entangled bits from victor.

I think it's kind of like looking at encrypted text from a one time pad. You can generate literally any plaintext from the encrypted text with some key, but without the original key, you can't recover the intended message.
posted by empath at 10:14 AM on April 25, 2012


then I don't see how QM experiments indicate that materialism is invalid

I don't see how experiments that have particles going back in time to kill their own grandfather square very well with a universe made of physical matter
posted by crayz at 10:53 AM on April 25, 2012


quantum steering into the past

You know, that sounds so sophisticated and diligent and responsible and promising and stuff.

You boys have fun. And get back to me when I can smell Lincoln's farts or something.
posted by Twang at 10:58 AM on April 25, 2012


I don't see how experiments that have particles going back in time to kill their own grandfather square very well with a universe made of physical matter

I'm sorry, but this doesn't have anything to do with grandfather paradox and doesn't violate it.
posted by empath at 11:14 AM on April 25, 2012


I'm sorry, but this doesn't have anything to do with grandfather paradox and doesn't violate it.

Tell that to Alice and Bob and their crazy [grand]son Victor, whose decisions alter the way Alice and Bob spun years ago
posted by crayz at 11:36 AM on April 25, 2012


They don't impact alice and bob, they impact the wavefunction that they recorded, but keep in mind that from the point of view of the photons, the measurements occurred simultaneously.
posted by empath at 11:44 AM on April 25, 2012


Carol and Ted declined to be interviewed.
posted by mmrtnt at 12:17 PM on April 25, 2012 [1 favorite]


Victor impacts the reality observed by Alice and Bob. Once you start imagining photons exist enough to have a point of view you're already lost in materialism. Other physicists at University of Vienna have done the double-slit experiment with molecules as large as 114 atoms, and that number is only going to increase over time. There's no reason why it would not be possible to do a double-slit experiment with an object large enough to be seen by the naked eye. It's worth pondering that for a while, I think
posted by crayz at 12:40 PM on April 25, 2012


Pushing the limits at best there - for macroscopic objects the de Broglie wavelength is generally tiny and you have trouble building suitable slits to diffract something through.
posted by edd at 12:52 PM on April 25, 2012 [1 favorite]


Once you start imagining photons exist enough to have a point of view you're already lost in materialism.

Are you claiming that photons are not physical?
posted by Bort at 12:55 PM on April 25, 2012


Are you claiming that photons are not physical?

Sorry - I should've seen you were the one to write "The idea that 'real physical objects' exist is one of our last, best myths" - so I guess you are saying that photons are not physical. I wish I'd realized that earlier and saved myself some time trying to engage you. :P
posted by Bort at 1:22 PM on April 25, 2012


Once you start imagining photons exist enough to have a point of view you're already lost in materialism.

I think you're all stuck in the metaphors, and you should focus on learning the math instead.
posted by empath at 5:01 PM on April 25, 2012


empath: From any subset of randomly generated bits you can choose many subsets which appear to be correlated or anti correlated. There is no way to distinguish those randomly correlated subsets from the actual subset that was correlated because of entanglement without the list of entangled bits from victor.

Having only read the Ars Technica article, it's still not clear to me why you couldn't modify the protocol to make this into a causality-violating signalling system.

Let's give Victor control over which measurement to make (either entangling or non-entangling), rather than making it a random decision. We'll repeat the experiment thousands of times (i.e. sending thousands of pairs of photons to Victor). Rather than varying it for each pair, Victor will make the same measurement (entangling or non-entangling) on each pair of photons he receives.

When Alice and Bob perform their thousands of pairs of measurements, they should have either thousands of entangled pairs of photons, or thousands of unentangled pairs of photons, and these two scenarios should be experimentally distinguishable. Thus they should be able to tell which measurement Victor (in the future) will be making.

This gives us what seems like a communication channel from Victor to Alice & Bob: if Victor wants to communicate a 0 to them, then he'll perform the non-entangling measurement on every pair that he receives; if he wants to communicate a 1, he'll perform the entangling measurement on every pair.

So why doesn't that work?
posted by logopetria at 3:55 AM on April 26, 2012


these two scenarios should be experimentally distinguishable.

How? That little bit 'experimentally distinguishable' is where the problem lies.

Suppose Victor entangles the pair. They could be entangled such that they must have the same polarisation, or they could be entangled such that they must have opposite polarisation and this will vary for every pair. Only Victor's measurement result tells you which of those two possibilities it must be. Only once you have Victor's result can you see that your result is consistent with it.
posted by edd at 4:20 AM on April 26, 2012 [1 favorite]


edd - ah, that was that point I hadn't picked up on, thanks!
posted by logopetria at 10:36 AM on April 26, 2012


Many Worlds Problem
posted by jeffburdges at 2:50 AM on May 1, 2012


« Older HowToBasic...  |  Beanplating on The Fifth Eleme... Newer »


This thread has been archived and is closed to new comments