# Entangled states before observation.

Tags:
1. Feb 7, 2016

I think this is just a quickie. I'm interested in what is assumed about entangled photons/particles before they are observed. Is it correct to assume that the photons/particles exist in all possible states simultaneously?
Thank you.

2. Feb 7, 2016

### Staff: Mentor

No.

QM is silent on what's going on when not observed.

Thanks
Bill

3. Feb 7, 2016

Thank you Bill,
by "silent" do you mean that QM assumes nothing at all about the nature of the photons separately before observaton or are you referring to properties related to entanglement only? I would have thought QM is not silent about several things for example it can assume that there are two photons each with the same and a known frequency. Or does the method of creating the entangled photons count as an observation?
Thank you.

4. Feb 7, 2016

### Staff: Mentor

Exactly.

In modern times generally an observation is assumed to occur just after decoherence ie in the assumption of an improper mixed state is a proper one is in effect an observation.

Thanks
Bill

5. Feb 7, 2016

Staff Emeritus
It's broader than that. Science doesn't answer the question "what is happening when we are not measuring". Science can tell you what the results of a measurement will be if you do it, but it can't tell you what the results of a measurement will be if you don't do it.

6. Feb 7, 2016

Thank you both but please consider my reference to, for example, frequency. Is it being suggested that in various experiments, such as quantum eraser experiments the frequencies of the photons are not known?

7. Feb 7, 2016

### Staff: Mentor

Its a frequency of a wavefunction and the theory is silent on if its real or not.

Thanks
Bill

8. Feb 7, 2016

Thank you for your replies. I'm not necessarily disagreeing with you but your answers seem to contradict information I have been getting from numerous other sources. Here's one example from the Kim et al paper referring to the output from a type 2 phase matching BBO crystal:

"A pair of 702.2nm orthogonally polarised signal idler photon is generated etc".

This is is to do with quantum theory and it seems to imply that the photons have properties about which the theory is not silent. Perhaps my researches have taken me to wrong or out of date places. It's a bit confusing at the moment so It would be helpful to get some references to put me on the right tracks.
Thank you.

9. Feb 7, 2016

### DrChinese

The referenced statement is accurate. The entangled outputs of a BBo crystal are something of a function of the output collection angles. The input photon goes straight in, but pairs do NOT come straight out. The deflection angle affects the pairings, and the frequency specifically IIRC. You can get pairs of, say, 700nm and 704nm if you so choose (and set up for this). The point is that they are specifically selecting on this attribute, and you should consider that along with their comment.

You are correct that there are a range of possible outputs generally, and they all obey a conservation principle.

10. Feb 7, 2016

### Staff: Mentor

Are you asking about entanglement here, or the more general notion of superposition? It's superposition, not entanglement, that raises the question of whether we should be thinking of the system as existing in more than one state at the same time. You can't have entanglement without superposition, but there are many situations in which a superposition can be described without considering entanglement.

Consider, for example the entangled two photon state used in so many discussions of Bell's theorem: $\frac{\sqrt{2}}{2}(|H\rangle|V\rangle+|V\rangle|H\rangle)$ - this is a superposition of the two states $|H\rangle|V\rangle$ (Alice's photon is horizontally polarized, Bob's is vertically polarized) and $|V\rangle|H\rangle$ (the other way around). It's also an entangled state because we cannot separate the state of one photon from the other; if a measurement finds Alice's photon to be vertically polarized we know that system state has collapsed to $|V\rangle|H\rangle$ so a measurement of Bob's photon will necessarily find it to be horizontally polarized (and vice versa).

But the entanglement has little to do with the question "Is it correct to assume that the photons/particles exist in all possible states simultaneously?" The two possible states making up the superposition are $|V\rangle|H\rangle$ and $|H\rangle|V\rangle$, and the answer to the question is the same as for any other superposition situation: "No", as Bhobba and V50 have said above. This is the same thinking that goes into the completely entanglement-free double-slit experiment, where the state of the particle is a superposition of "left-slit" and "right-slit". As in that experiment, any attempt to read more into the state than that, to say that it is really one or the other but we don't know which, or that it is both at the same time, will run into various logical difficulties and contradictions.

It is correct to say that the system is in the state $\frac{\sqrt{2}}{2}(|H\rangle|V\rangle+|V\rangle|H\rangle)$ and it is correct to say that there is a 50% chance of finding Alice's photon vertically polarized and Bob's photon horizontally polarized and a 50% chance of finding Alice's photon horizontally polarized and Bob's photon vertically polarized (and that leaves 0% for any other possibility). But that's about all that you can take to the bank here.

Last edited: Feb 7, 2016
11. Feb 7, 2016

Thank you all for your replies. I know there are various interpretations of quantum theory but basically I'm trying to find out if there is a concensus of opinion about what properties, if any, do photons, for example, have when they are not observed. For example I think it's generally accepted that each photon has a certain speed and a certain frequency/ wavelength.
It seems I wasn't clear in my opening question so please allow me to rephrase the second sentence:
Is it correct to assume that before observation the photons/particles exist in a superposition of all possible states simultaneously? I guess the answer is still no.
Thank you

12. Feb 7, 2016

### Heinera

Before observation, we don't even know if the photons/particles exist at all. By "superposition", we simply mean a formal mechanism that produces a statistical prediction for the result of a possible measurement. It says nothing about how things "exist" before the measurement is performed.
Yes.

13. Feb 7, 2016

Thank you. Most of this is making some sense except I keep reading apparently contradictory comments from other sources. For just one example please look again at the quote I gave in post 9 which referred to the quantum eraser experiment of Kim et al.
I'm not disputing any comments that have been written here but I'm getting confused by the apparent lack of agreement between several of the different sources I have looked at whilst researching this question.

14. Feb 7, 2016

### Staff: Mentor

15. Feb 8, 2016

### vanhees71

No, it's assumed that the photons are in the entangled state. It is wrong to say that just, because the state is a superposition in some arbitrary basis, it has the properties described by the basis at once. It simply doesn't make sense and contradicts Born's rule, which tells you the probabilistic meaning of states.

16. Feb 8, 2016

Thank you everybody for your inputs. Bhobba, I had previously looked at one of the papers you referred to in the earlier thread- the Thomas V Marcella paper which is at an "intermediate" level. The paper summarises the main difficulty which I and I guess many other people are experiencing. I quote:

"Numerous text books and journal articles discuss slit interference usually in conjunction with wave particle duality."

That reveals the problem. We read something in one paper and then read something seemingly contradictory in a different paper. Who is right and who is wrong? Could it be that different teams have different ways of analysing the same problems and are some methods better than others? There are so many questions.

Personally I have my own views which are in broad agreement with most of the comments that have been made in this thread. I just wanted to know if there was general agreement amongst the wider community.

Thank you everyone.

17. Feb 9, 2016

### vanhees71

Wave-particle duality is an idea from "old quantum theory", which is known to be much less consistent than "modern quantum theory", which was discovered in 1925. The problem is that many textbooks use the historical approach to teach QT and start with explaining the "old quantum theory" at length. In my opinion this only adds to the confusion of learners of QT. In modern QT there is no wave-particle duality but only the description of quanta in terms of a quite formal mathematical structure. The link to physics is due to the Born rule and thus a probabilistic meaning of the quantum states. As soon as one uses this minimal interpretation, all apparent paradoxa are gone. In other words: QT tells us that nature doesn't behave as classical particles or classical fields but in terms of QT. That's it.

18. Feb 9, 2016

### Staff: Mentor

Thanks
Bill

Last edited: Feb 9, 2016
19. Feb 9, 2016

https://www.physicsforums.com/https:www.sciencenews.org/blog/context/new-analysis-rescues-wave-particle-duality [Broken].

Thank you. I am convinced that modern QT is very succesful and I'm not disputing that one little bit. However, if you do a quick search you will find that concepts such as wave particle duality are referred to in the newer literature as well as in the old. I'm not getting this information from textbooks but from journals and other sources I'm finding on the net.

The work reported in the link added above was published just recently and it was one of the first things I came across when I googled wave particle duality. It might not be a reliable account but it's an example of what's going on. The quantum eraser experiment I referred to above mentions wave particle duality.See paragraph 1 page2.
I quote:

"This reflects the wave properties (both path) of photon 1 ............................................................................. we have now detected the the particle property (which path) of photon 1".

I'm not trying to promote the concept of wave particle duality or any other concepts that might seem to be weird or paradoxical. In fact the opposite is the case and I want to know why there are so many apparently contradictory views.

Last edited by a moderator: May 7, 2017
20. Feb 9, 2016

### Staff: Mentor

I explained why. Precisely what don't you get about the link I gave?

The concepts in QM are such that they need to be built up to.

If you want the real deal read Ballentine - but as you will find if you do, its very hard without knowing the books that lead to it. Even Dirac's famous book mostly does it correctly and its decades old - yet concepts it eschews still appear all over the place. The same with Von-Neumann's classic. Many ideas long superseded and gotten rid of ages ago are still about. Most don't worry about it because the context is understood and most aren't worried about foundational issues anyway. But its central to many that post here - and that's why confusion occurs.

Some people aren't that interested in the foundations of QM and don't proceed to a book like Ballentine that clearly explains the basic concepts and supersedes the less advanced texts. Its not some great 'swindle' - its simply there are different levels of understanding.

Thanks
Bill

Last edited: Feb 9, 2016
21. Feb 9, 2016

I don't think you understand the point I'm making. Let me put it differently. According to the discussion here it seems there are two broad camps when it comes to concepts that might seem weird. For the sake of brevity let me refer again to just one of those concepts-duality

1. In one camp there are those who follow what has been described here as modern QM and who have no need of duality.
2. Opposite to that there are those to whom duality is a concept which needs to be considered.

I lean more towards camp 2 but that's beside the point. What I want to know is why there is so much literature by people who appear to be in camp 2. This is not just the old literature either, its new literature as well and much of it published in journals. Why do the journals publish if it is wrong or if it displays a lack of understanding?
The problem I'm getting in this thread is that mainly I'm getting comments from people in camp 1 only, I'm not hearing anything from the other side. Perhaps members of camp 2 might have something valid to say.

22. Feb 9, 2016

### Staff: Mentor

What I am saying is camp 2 is simply people who most likely know better but also know it doesn't affect shut up and calculate, or are not particularly concerned with it to read the more advanced foundational texts because it doesn't affect shut up and calculate.

A number of practitioners who post here have commented we get far to caught up in this stuff. The reason is a lot of people that post are really caught up in foundational stuff but those that practice it aren't.

Thanks
Bill

23. Feb 9, 2016

### Staff: Mentor

Then you should read books like Ballentine - or even Dirac's ancient classic. If you don't want to because it's a slog you have your reason why camp 2 flourishes.

Thanks
Bill

24. Feb 9, 2016

### Staff: Mentor

One possible explanation for why you're only hearing from people in camp one would be that camp two is less well populated than you're thinking.

And seriously, kidding aside, the disagreements you're seeing are (mostly) illusions caused by using natural language to describe mathematical concepts. The mathematical formalism of quantum mechanics is clear, unambiguous, and contains nothing that looks like the popular notion of "wave-particle duality", and that's what quantum mechanics has been for better than three-quarters of a century now. However, when we try talking about it without the math:
1) Sometimes people who know what they're talking are careless with the English words wrapped around the math. They do this because the truth is in the math and not the words, so any ambiguity and imprecision in the words is harmless because the math is what counts. Unfortunately, someone looking at the words without the math background or someone to help with further explanation is going to be misled.
2) Often, it's just plain hard to do it right. Some popular misconceptions so deeply embedded (due, as @vanhees71 pointed out above, to the tradition of explaining QM from the historical perspective) in the popular mind that it's easier to explain around them. If you only have a few paragraphs or a few pages to give a math-free overview of an experiment... Are you going to start with "before you can understand this, you'll have to forget everything you think you know, and then spend a few evenings working through the first two hundred pages of Ghiradi's 'Sneaking a peek at God's cards'" or are you going to start with "You've heard how things can be particles or waves? Well, in this experiment that means....."?

A forum like this one has the advantage of being interactive so we can work for a better level of understanding; that's what's behind @bhobba saying "You generally get the real deal here - not watered down popularisations of dubious validity.".

25. Feb 9, 2016

### vanhees71

I can only cite Einstein again: "Make things as simple as possible but not simpler." The most difficult thing is to write good popular-science books on highly abstract subjects like quantum theory. In my way it's nearly impossible to write one that's not oversimplified to the extent that it becomes wrong. There is no way to describe quantum theory accurately without a minimum of quite tough mathematics, including functional analysis of Hilbert spaces and probability theory.

The old fasihioned quantum theory before 1925 has never been a consistent model, and this was well known by the great physicists who discovered them, among them Planck (who was much closer to the modern point of view on a intuitive level than, e.g., Einstein, but he was never satisfied with modern quantum theory either), Einstein, de Broglie, Bohr, and Sommerfeld. To a certain extent the discovery of modern quantum theory occured because of the deeply dissatisfying state of atomic physics in terms of the Bohr-Sommerfeld tradition. Heisenberg was one of the master pupils of Sommerfeld's and from day 1 in the institute involved in atomic physics in the old-fashioned quantum theory. So it is no surprise that Heisenberg found the modern quantum theory first on his famous escape from sever hay-fever attacks to the island of Helgoland in the summer of 1925. As was very quickly worked out by Born, Jordan, and Heisenberg this was quantum theory with the Hilbert-space structure in terms of the representation of Hilbert space as $\ell^2$, the Hilbert space of square-summable sequences. The operators are infinitely-dimensional matrices, and that's why this formulation is known as "matrix mechanics". One early breakthrough was the ingenious solution of the hydrogen problem (quantum mechanical "Kepler problem") by Pauli, using the O(4) dynamical symmetry. A bit later Schrödinger worked out "wave mechanics", which turned out to be much more convenient for many praktical problems than "marix mechanics". It was Schrödinger himself who showed that both theories are in fact completely equivalent, and also around the same time Dirac came up with the representation-independent formulation we still use (and also should use in teaching the subject from the very beginning). Last but not least John von Neumann brought the theory in a mathematically rigid form, including a mathematically strict treatment of the eigenvalue problem of unbound operators on Hilbert space.

It is precisely this historical development that should be taught after one has introduced the modern QT (either as wave mechanics or better from the very start in the representation-free formulation), so that the old-fashioned quantum theory cannot lead to wrong intuitive pictures, you have to unlearn before you can open yourself to the modern point of view. At least that's the experience I made in my learning history of QT: In high school we got quite well trained in the old-fashioned ideas, and we also learnt some simple wave mechanics. Later at the university, I had to unlearn the old-fashioned QT pictures ("wave-particle dualism", "Bohr-Sommerfeld orbits", "correspondence principle" and "complemenarity", which was invented by Bohr in connection with modern QT but is a pretty empty metaphysical concept).