Talk:Quantum entanglement

(13 Feb 2003) Though I have read the discussion below about transmitting (and not transmitting) information, I am afraid I still don't "get it." Perhaps my brain stumbles on the meaning of "classical" information. To my average mind, "information" would include the fact that the state of a paired photon had changed. If that sort of "information" is detectable, it seems fairly easy to imagine that such transmissions could be scaled up and such transaction could be used to transmit any sort of "information."

What raised the issue is the latest of several New Scientist articles, http://www.newscientist.com/news/news.jsp?id=ns99993384, mentioning entanglement being used to transmit information. I confess I don't fully understand the articles.

In quantum mechanics, the only "information" you can obtain is by performing a measurement. In this case, this is the measurement of "spin". Nothing else can tell you anything about the system, including whether or not "the state of a paired photon had changed." As for quantum teleportation, see that article. -- CYD

---

I corrected a small confusion between the definitions of 'pure' and 'separable' states. [from the Centre for Quantum Computation, Oxford]

Do you have a reference? Several books and webpages I have read, on QC and other topics, simply refer to it as a "pure" (as opposed to "mixed") state. For example, Sakurai refers to a "pure ensemble". -- CYD

Hi. Indeed pure states are opposed to mixed states. But both types can also be either separable or not, and that was the confusion that I tried to correct. If a bipartite state (pure or mixed) is separable then its parts can be produced separately by classically correlated sources. If it is not separable, then the state is said to be entangled, and can exhibit non-locality, contextuality and such counter-intuitive phenomena. A review on separability can be found on http://uk.arxiv.org/abs/quant-ph/0006064 but it is quite technical. I again read through and corrected a few problems arising from this confusion.

---

claim that hidden variables theories can't be local removed

Would you care to comment on why this claim is inaccurate? It seems to me a cogent description of the outcome of Bell's theorem. -- CYD

Then Bell's theorem is wrong.

Many-worlds is a counter-example to the claim. It is a purely local theory which is also deterministic. It is also the quintessence of a hidden variables theory; think of all those worlds which we can never observe. I could be wrong about the locality of many-worlds but most people seem to consider it local in the important sense (that non-local effects are unnecessary to explain the Aspect experiment). -- ark

Many-worlds is local, but it is not a hidden-variable theory in the sense of Bell's theorem. I think the characterization of Bell's theorem was correct. See EPR paradox. AxelBoldt, Thursday, May 30, 2002

I don't think it makes sense to talk about many-worlds being "local", because the Hilbert space of a system need not be the space of square integrable wavefunctions. If I'm not mistaken, that's how non-local phenomena (such as the EPR paradox) come about. -- CYD

Whatever you guys decide, the EPR paradox page needs to seriously change. For example, it says "many hidden variables theories have been constructed" and then says "but experiments sided with QM against them". What theories are they talking about? I don't know and it will seem to most people that it's talking about Many-worlds. That's just one example by the way.

Also, an alternative to interpretating Bell's theorem as saying that Many-worlds is not a hidden-variable theory, is simply to say that Bell's theorem is meaningless because it uses a concept of "hidden-variable" which ultimately interests no one. -- ark

Well, obviously that kind of hidden-variable theory interests no one now. On the other hand, EP&R did suggest that quantum mechanics was incomplete, relying on some underlying classical hidden variables. I'm not actually sure if their suggestion was developed by anyone into a workable theory, but Bell's theorem imposes quite strong constraints on the kind of hidden variable theory that can be formulated while matching reality. -- CYD

Something that physicists (and serious physics students) seem to rarely understand is that the mathematics of physics is a completely different subject from the history of physics, which is still different from the ontology of physics (what exists in reality, what I would call the essence of physics).

And they never understand that you can and should teach these three very different subjects separately. So if Bell's inequality is only of historical importance then it's useless and should never be mentioned outside a class on historical physics, where it would be taught aside other great failures of physics. If there's anything in Bell's analysis that's actually relevant to the ontologies of modern physics theories, then that should be carefully extracted out and the rest discarded as chaff. -- ark

---

"Entanglement obeys the letter if not the spirit of relativity"

It's funny to read this given:

The peculiar aspects of quantum spin measurements in EPR-type experiments can be regarded as a natural extension of the principle of special relativity. (http://www.mathpages.com/rr/s9-09/9-09.htm)

The "spirit of relativity" refers to the locality principle, which it's fairly clear was what EP&R were concerned about. The link you provided offers an analogy between quantum measurements and relativity - which is not incorrect, but it's not the standard approach to this problem, AFAIK. -- CYD

Of course, the standard approaches don't seem to be a lot of help to most people, especially non-experts, in understanding physics. So I think that, like other ambiguous stuff that you need to be a physicist (and preferably a historian of physics) to understand, it should just be removed. (Like I did on the Copenhagen page. What do you think of it btw?) -- ark

---

Someone deleted my previous note that at least one scientist has actually used this to transmit information, based on the "fact" that its not possible to do so. Tell that to him--he'd already done it. Classical music, to my recollection. Anyway, here is a formal piece of evidence thats just cropped up:

http://www.theaustralian.news.com.au/common/story_page/0,5744,4522247%255E2702,00

The URL's a little mangled...

--Alan D

Actually, all they did was quantum teleportation, which never transmits classical information faster than the speed of light, and isn't even helpful for sending classical information. One key paragraph is:

An encoded radio signal is embedded on an input laser, which is combined with entanglement and then scanned. The laser is destroyed in the process. But the radio signal survives and is sent electronically to a receiving station, where within a nanosecond an exact replica of the beam - with the radio signal intact - is retrieved and decoded.

Notice that in order to teleport the laser from point A to point B, they had to first destroy the beam, then send radio waves (at the speed of light) from point A to B, then recreate the beam. You could say that the quantum state was teleported instantly, but the people at B couldn't detect that it had been teleported until after the slow radio waves reached them. Point B didn't end up with any more classical information than was already present in those radio waves. Quantum teleportation has been done by a number of people now, but it has never sent a single bit of classical information. -LC, Sunday, June 16, 2002

OK, I misinterpreted that part...I'm certainly no expert :-) I'm wondering now if the thing on the classical music was the same. He explicitly stated however that he was transmitting information using the technique, but it could have been sensationalistic journalism. I wish I could remember where I saw it.

--alan d

If I'm not mistaken, the reference you provided covers a technique called quantum teleportation, which is not what this article is talking about. The article is noting that it is impossible to transfer information using the EPR setup; otherwise, you would be able to send information faster than light. It is possible to transfer information using quantum teleportation, but then the information will not travel faster than light. -- CYD

---

I find the first sentence of this article potentially highly misleading: The entanglement of two quantum systems that are far apart does NOT lead to them interacting with each other over the distance - at least not within the standard interpretation of quantum mechanics. They just may show correlations that are so strong that, if you want to explain them in terms of a hidden-variable theory (as EPR intended to, for example), this hidden-variable theory cannot be a local theory. For example, in Bohm's h.v. theory (which agrees with q.m. predictions and therefore, according to Bell, has to be nonlocal) one can indeed see how one of the particles is influenced by what the other one does (which may be influenced by the setting of the other detector). But as you point out yourself, it cannot be used for superluminal communication, because the outcomes are statistical, and on the average there is no mutual influence. Since usual QM only talks about statistical results anyway, it is misleading (at least for the beginner) to imagine a superluminal influence appearing between the particles.

How about something like this: Entanglement is a purely quantum-mechanical property of two particles whose physical attributes (like position, momentum or spin) have become correlated after interacting with each other. The correlations between such "entangled" particles may be shown to be stronger than any "classical-like" theory allows, under appropriate circumstances (see below for more details). -- FlorianMarquardt

I went ahead and did something like this. Feel free to make such changes without posting to the Talk page, as that is more efficient. Welcome to Wikipedia, by the way! -- CYD