103 Lemmel and Hoffman experiment with reality

[00:00:00] Holger Hofmann: …and you are trying to somehow fit your idea of reality, which is not based on observation into the observable universe. For me, I’m always wondering how can I have confidence about such a thing when my only confidence about reality in everyday life is the observation?

[00:00:22] Al Scott: The Rational View is a weekly series hosted by me, Dr. Alan Scott, providing a rational, evidence-based perspective on important societal issues.

[00:00:37] Soapbox Media LLC: Produced by Soapbox Media.

[00:00:42] Al Scott: Hello, and welcome to another episode of The Rational View. I’m your host, Dr. Al Scott.

[00:00:48] Quantum Mechanics is a strange theory. Dr. Richard Feynman said, “I think it is safe to say nobody understands quantum mechanics.” [00:01:00] So why is this theory so popular if nobody even understands it? How is this the basis for all of our physical knowledge of particles and their interactions? Well, quantum mechanics, what I’ve learned going through physics as an undergrad and a graduate student, quantum mechanics did not come from a philosophical, deep understanding of the structure of reality. It came from observations made in repeatable experiments on particles. And these observations could not be explained by a classical means. By assuming that the particles had a reality when they weren’t being observed. And that’s horribly difficult to understand. For a long time, physicists were unable to decide if fundamental particles acted like billiard balls or like waves and neutron, scattering experiments showed that these objects acted like tiny billiard balls or, nuclear scattering I should say. [00:02:00] This duality between acting like billiard balls and waves was unlike anything anyone had seen in a macroscopic sense. There’s no good analogies in everyday life to this behavior.

[00:02:13] The strangeness of quantum mechanics was best seen in the dual slit experiment, where you direct a stream of sub atomic particles at a screen with two slits. And instead of getting two hills on a detector on a piece of film behind those slits, what you see is an interference pattern of peaks and valleys as though the particles were waves being pushed through that pair of slits. But when they hit the screen, they showed up as dots. They were again, back to billiard balls, tiny billiard balls.

[00:02:49] What does it mean? Physicists have found that if you can find out enough to distinguish between the paths that a particle takes, the interference pattern disappears. If you look [00:03:00] at the particles as they’re going through, if you tag them in some way, they no longer interfere, they become distinguishable. But if they’re indistinguishable, you get interference. And this means that we don’t know what’s going on from analogy. What does it mean? But a century ago we found that the Schrödinger equation of quantum mechanics accurately describes the progress of the wave function of any system of particles through any experiment, and it can be used to determine the probability distribution of large numbers of particles going through repeated interactions. But it said nothing about how an individual particle goes through. It said nothing about how particles turned from waves into particles or when they were each different. It was up to the physicists to interpret this, to get a consistent interpretation of the math. And this was coined as the measurement problem. What aspect of a measurement collapses this [00:04:00] distributed wave function to a single point like interaction. Nobody knew there are no arbitrary constraints on this interaction. This is just something that physicists say, “okay, we’ve observed it now. It’s here. The wave function has collapsed.” But what aspect of the measurement collapses the wave function? Do you need to be an intelligent observer to do this? Do you just need an atomic interaction? An energy sharing interaction? Nobody knew what were the possible interpretations, many physicists feel uncomfortable dealing with the philosophical implications of this problem. The only thing we do know is that the theory, as far as it goes, is unequivocally, unquestionably correct. It’s the most highly verified physical theory that we have, and it’s been measured to something like 23 orders of magnitude to be correct through very precise experiments. And there’s been several different interpretations. The Copenhagen interpretation, [00:05:00] you know, a hundred years ago, got us the basic idea that the wave function is just a probability distribution. And we, by calculating the, the magnitude of this wave function and integrating over your measurement space, then you get the probability that a particle will end up there, and that’s all it says. But there are other interpretations that try to give reality to the particles. The Many Worlds Theorem says the wave function never collapses and the particles have basic reality, as far as I understand, and making a measurement doesn’t collapse the wave function, but it tells you which of an infinite number of universes you’re actually in. And until you make that measurement, you have this probability distribution of universes that you might exist in. And all these universes are overlapping and interfering.

[00:05:47] A recent innovative experiment, fires neutrons through a double slit and proves that each neutron goes through both slits at the same [00:06:00] time.

[00:06:01] If you like what you’re hearing, please press like on your podcast app, share it with your friends or join us to discuss these experiments on my Facebook group; The Rational View.

[00:06:14] Professor Holger Hoffman studied physics in his hometown of Stuttgart, Germany, then went to Tokyo University for a post-doctoral Fellowship in 1999. Now at Hiroshima university his research is focused on the way quantum theory describes observable phenomena. He believes that the key to a proper understanding of quantum mechanics can be found by exploring the practical means that we need to employ to achieve optimal control over a physical system. Dr. Hoffman, welcome to The Rational View.

[00:06:47] Holger Hofmann: Thank you.

[00:06:50] Al Scott: Dr. Hartmut Lemmel studied Physics in his hometown, Vienna and graduated at the Vienna University of Technology in the group of Helmouth Roche, who is the father of [00:07:00] neutron interferometry. Dr. Lemmel welcome to The Rational View.

[00:07:04] Hartmut Lemmel: Thank you.

[00:07:07] Al Scott: So together you have recently published what I consider to be a groundbreaking paper called Quantifying the Presence of a Neutron in the Paths of an Interferometer. And I just wanna read the abstract and then we can discuss. So the abstract says, “it is commonly assumed that no accurate experimental information can be obtained on the path taken by a particle when quantum interference between the paths is observed. However, recent progress in the measurement and control of quantum systems may provide the missing information by circumventing the conventional uncertainty limits. Here we experimentally investigate the possibility that an individual neutron moving through a two path interferometer may actually be physically distributed between the two paths. For this purpose, it is important to distinguish between the probability of finding the complete particle in one of the [00:08:00] paths and the distribution of an individual particle over both paths. We accomplished this distinction by applying a magnetic field in only one of the paths and observing the exact value of its effect on the neutron spin in the two output ports of the interferometer. The results show that individual particles experience a specific fraction of the magnetic field applied in one of the paths, indicating that a fraction or even a multiple of the particle was present in the path before the interference of the two paths was registered. The obtained path presence equals the weak value of the path projector and is not a statistical average, but applies to every individual neutron verified by the recently introduced method of feedback compensation.”

[00:08:46] Wow. Holger, could you provide some background and what motivated this particular experiment?

[00:08:53] Holger Hofmann: Okay. Yeah. Well, there’s a lot of background here and thanks for reading the whole abstract, it’s [00:09:00] quite a lot, right? That’s been an old problem, really; what actually happens between sending in a particle into an interferometer and detecting it at the output, and the conventional wisdom is really that we cannot tell. So if you go back to the Copenhagen Interpretation, it’s basically just uncertainty. And if you measure the interference, you really know nothing about the path of the particle. But there has been some progress here, especially with the idea that you could make an interaction so weak that the interference doesn’t disappear and then maybe you could get a signal out. And on the average, actually you would get the values that we also get. But there has been a big controversy because that statistics and the noise is huge. And this new method that, actually I introduced basically just one year before, this new method using a feedback basically confirms [00:10:00] that this weak value is completely accurate and has no errors.

[00:10:06] Now, the reason why this is possible is that we focus on the amount of change that the system obtains. In this case, that’s the rotation of the spin. And if you look at the details of the paper, the spin of the neutron is really the probe. It gets rotated. And the normal problem is that the spin is so “quantum noisy;” so uncertain that if you just look at the spin you don’t see much information. But what I found is that the trick is really that if you compensate the rotation and then look in the direction where the spin is maximal, basically detected at 100%, always in the same direction. To do that, you need the information; how big the angle of rotation is. That’s why it doesn’t work in the [00:11:00] forward direction. There’s this old paradox that if you could measure where the particle is before the interference happened, then you would normally argue that even if the interference doesn’t happen, the same result must come out. So suppose that afterwards, you just decide to measure which part the particle is in, we know that if you directly detect the particle in the past, you find it or you don’t find it. That’s just two possibilities. So the big paradox is when you do interference, how come the particle can be distributed?

[00:11:37] A lot of people would call that a paradox and even dismiss the possibility. But what we find is that the rotation angle is really determined in this fractional way. It’s just that you cannot use the small rotation to get the information.

[00:11:58] Al Scott: So what you’re doing is, you [00:12:00] have an interferometer, which has two paths that the neutron could choose randomly, thinking about it in a classical way. And in one of these paths, there’s a magnetic field that rotates the spin of the neutron, a little bit. And so, from when I learned physics, it said if there’s any which way information on an interferometer it would destroy the interference pattern; it would just choose one of the two paths and you’d no longer have interference. So what’s happening here is that you’re somehow providing some which way information and not destroying the interference.

[00:12:32] Can you, can you explain how that works?

[00:12:35] Holger Hofmann: The information has to be really small. The trick here is we are not using the information on the spin rotation to determine the path. We are determining the interference outcome, and then we make conclusion from the interference outcome, what the path should be, and reverse the rotation. So the point is that we can [00:13:00] accurately test whether we have estimated the right rotation angle. Even so, we cannot read out the rotation angle in the forward direction. So this is actually a very tricky business with a difference between extracting information and verifying a hypothesis. So it’s a little bit tricky because I mean, you would normally say, I mean, this is not a measurement. The spin is not measuring the path. You cannot see the path by looking at the spin. But you can confirm the presence in the path by testing the spin.

[00:13:45] Al Scott: Hmm, and the measurement implies that you don’t get some neutrons with the full rotation, and some without. What you get is a fraction of the rotation in all of the neutrons. [00:14:00] And this means that the neutron has split itself between these paths, effectively. Just like the wave function does. So the wave function splits up, and then the question is, what’s the interpretation? And you have some interpretations that there’s a, a pilot wave guiding us a real small particle through one or the other of the paths, or you have other ideas that you know, you have multiple worlds and, and the measurement tells you which world you’re in, and in one world that particle took this path and the other world, the particle took the other path and these worlds interfere. So you’re basically showing… doing an experiment, which I think actually tests these interpretations. Is that the idea behind this?

[00:14:43] Holger Hofmann: I wouldn’t say that’s the idea behind that. But it effectively is like that. For my side, I can say that my whole interest here comes from what I said in the introduction actually. I really don’t think that we can make a meaningful argument about interpretations [00:15:00] without considering what we are talking about in terms of the evidence. So you need to actually have an effect. And in fact, when we wrote the paper and we had a long discussion about this problem; what is the presence of a particle? And the problem here is that we all imagine that we know what the presence of a particle is, but in reality this knowledge comes from thinking we can see it. Thinking we can touch it. So if the particle cannot be seen or touched, it’s actually a really difficult question what we mean by the presence of the particle. So that’s, I think is the, the big problem, for example, with the guiding wave theory, that these particles do not actually have any other role, except that you claim that when you detect it at the end it’s because it traveled along a path that nobody ever saw.

[00:15:55] So we are imagining that we could make ourselves really small and look, where’s the [00:16:00] particle. But there is no physics to that. You can’t do that.

[00:16:04] Al Scott: This seems to be like the overlap of physics and philosophy.

[00:16:08] Holger Hofmann: Yeah, but it’s rather important because a physicist also has to know what they’re doing. And the danger is that we are actually using language in the wrong way.

[00:16:17] This has been overlooked, I think, in the very beginning of quantum mechanics. There’s actually an interesting piece of writing by Heisenberg, where Heisenberg basically explains that quantum mechanics constitutes an abuse of language, but that cannot be avoided because the classical language does not work.

[00:16:36] Al Scott: Yeah, there is no classical analog to this situation that… you can’t get a good feeling for what the heck is going on.

[00:16:44] Holger Hofmann: Right. There is no microscopic experience, right? Okay. Maybe Hartmut wants to say something about that. Also it’d be interesting because of course, Hartmut is closer to the actual experiment and deals with what is happening there.

[00:16:56] Hartmut Lemmel: Yeah, well, I’m say the experimentalist of this [00:17:00] work, and in my group we’ve been doing neutron interferometry for a few decades already. And in the neutron interferometer, the neutron is split into two paths and then the neutrons recombine, but it is done not with like laser waves, but it’s with neutrons, which are massive particles. And if you look at the velocity of the neutrons and take into account the size et cetera, you conclude that every neutron takes a 10th of a microsecond to pass the apparatus. But the intensity is so low that only every microsecond there is one neutron on average. So you see, there are never two neutrons simultaneously in the apparatus and still an interference pattern builds up. So it is a clear sign that it is single particle interference. It is self interference. Each particle somehow feels both waves, both ways. So our finding is actually, in my feeling, I was not surprised. This is what I [00:18:00] always imagined. And if we talk about the pilot wave or interpretations like that in my opinion, if you say every neutron is accompanied by a pilot wave, then it’s an integral part of this party.

[00:18:17] So if you say, okay, the neutron itself goes only one way, but the pilot wave goes both ways, I think you haven’t won anything. It’s a semantics question.

[00:18:25] Holger Hofmann: I’m very tempted here to ask… I should know, but I would like to point out and ask about the final detection of the neutron. I mean, the neutron is actually detected in the end, in a nuclear reaction as far as I understood. That is, yeah, that’s correct. So here’s the tricky part for the pilot wave theory, because basically the pilot wave theory drops the use of the wave completely at the moment of this detection. Mm-hmm so there’s a little bit of a strange idea that this kind of interaction, this nuclear reaction at [00:19:00] the end is completely different in nature from the quantum mechanics we used before. Of course I guess that in the experiment, the control of this process is not very precise, right?

[00:19:13] Hartmut Lemmel: Well, there’s a certain volume, it’s a gas cell where the, the neutron can be absorbed and they then send out which are actually then detected. And yeah, in principle, one can find out where this happened exactly.

[00:19:31] Holger Hofmann: Up to a certain position, which as a theorist, I would say the position is not that high because if I would calculate the interferences that are involved here and there are interference that I mean, I would need information that I never get in the experiment in this case.

[00:19:46] Hartmut Lemmel: Well, I mean, in the neutron interferometer, we don’t need a very higher resolution because we have only two exit beams, but there exists of course also neutron detectors with spatial resolution and time resolution. And of course, it’s a question of [00:20:00] technology and statistics what occurs; you get out of it.

[00:20:04] Al Scott: Maybe you could walk us through a little bit more through the hardware you need to make this measurement. How is this experiment set up, Hartmut?

[00:20:12] Hartmut Lemmel: So the neutron interferometer consists of a perfect silicon crystal. So that’s a crystal about 20 centimeters long and there are three lamella standing up out of a common crystal base and each lamella is a beam splitter. So this is based on, on fraction. The neutron beam with the right angle and the right wavelength is deflected by the lattice of the crystal. And if you do it right, then you have 50/ 50 chance that the neutron is transmitted or reflected. And this is then again, the middle lamella where the neutron beam is reflected again, towards each other. Then they leave the interferometer set up in different directions. And then of the meter way or whatever [00:21:00] space you need, you put the neutron detector and detect the neutrons and in between, inside interferometer, you have space to place spin flippers or spin, rotators, absorbers, whatever you like. And yeah, although it’s every single neutron, which is going these two paths, the paths are macroscopically separated by in the order of centimeters.

[00:21:28] Al Scott: So you have two different paths centimeters apart, and what’s the wavelength of a neutron? What’s, you know, a wave packet that represents the neutron?

[00:21:36] Hartmut Lemmel: That’s two angstrom. So two times 10 to the minus 10 meter, that’s really small. That’s about the distance of two atoms in the Silicon crystal.

[00:21:46] Al Scott: Right, so the intrinsic size.

[00:21:48] Hartmut Lemmel: Very small. And that’s why the parameter is very sensitive to vibrations. For example, if you move one of these crystal lamella by one lattice constant, you would [00:22:00] introduce already a phase shift of two pi. And so you have to control the relative position between these lamella really good. That’s why they are made monolithically out of, they’re cut out of a single crystal in which they’re still physically connected, so that they’re also kept in position. Always kept in position.

[00:22:22] Al Scott: So this is one big Silicon crystal that has various interaction zones in it?

[00:22:28] Hartmut Lemmel: Mm-hmm right.

[00:22:29] Al Scott: Wow. And then you have to basically get rid of vibrations at the angstrom level.

[00:22:36] Hartmut Lemmel: Exactly. This sits on an optical table, which dampens the vibrations, the temperature should be uniform. It should not change because then the lattice constant changes, et cetera.

[00:22:48] Al Scott: How long did it take to set this up to make these measurements?

[00:22:53] Hartmut Lemmel: This particular experiment, for example, I think we had beam time of two weeks and approximately [00:23:00] it was 10 days alignment and then a few days data taken.

[00:23:04] Al Scott: Wow. That’s impressive. It’s actually very quick.

[00:23:08] Holger Hofmann: Yeah, it should be mentioned it’s much quicker than we are in optics, but the reason is that, I mean Hartmut and the guys are very well set up really. So the equipment is already present.

[00:23:19] Hartmut Lemmel: That’s it, we didn’t have to develop new components or so use what we had.

[00:23:27] Al Scott: Interesting. So, were you surprised at the results or was this what you expected to see?

[00:23:33] Hartmut Lemmel: Yeah, I expected it, but I was surprised that the theory can make this statement that it really applies to every single neutron.

[00:23:45] Holger Hofmann: Right, right. And I wasn’t surprised because the theory made me quite confident that this would be to the side. But I have to say, when I discovered the theory, I was a bit surprised myself that this is possible. Right. [00:24:00] So mm-hmm, the essential point is really that before I too had thought of these spin rotations are something that you cannot directly observe or measure, and I hadn’t thought of the possibility of compensating the rotation and confirming that nothing happened. That was really a new idea. And that changed the whole situation a bit.

[00:24:22] Al Scott: Right. So you’re applying a compensation after they’ve recombined?

[00:24:26] Holger Hofmann: Yes. Yes. After the interference. So the trick here is that actually the theory also says that there’s a quantum interference effect involved that actually changes the angle of rotation depending on which balance of interference you have. It’s a little bit complicated. So this is precisely where quantum mechanics is quite different from classical physics.

[00:24:50] When I did the theory, and that’s something that has not been considered enough, if you interfere whole operations, you can sometimes get a new operation completely [00:25:00] out of two operations that were completely different, right? So it’s not the same as interference of waves. It’s really an interference of operations.

[00:25:11] Or so the story, what happened is actually quite flexible, right? And this is what happens to the spin rotation. So if you look at the spin rotation in the one pass, it’s this alpha rotation and the other pass, there is no rotation. So you think these are the two possibilities, but after the interference, you actually get new possibilities.

[00:25:33] Now, you shouldn’t say that it’s after the interference, actually, because before the interference, you simply have ambiguity. In the end and after the interference, you could undo the interference. Then you had to go back to ambiguity. So this is where you have to go all the way to the final measurement.

[00:25:52] And that’s the confusing thing in the end. This is where this thing comes in for me with the [00:26:00] nuclear reaction at the very end, right? That’s where things are finally pinned down because before that, all the interactions were too weak to establish their own reality. And that’s a little bit different.

[00:26:15] This is actually where quantum theory comes into its own, in my opinion, right. Because when you imagine forces in classical physics, a force pushes something and there’s a change, and the change is a fact. In quantum mechanics there is absolutely no description of a change where you can then identify the amount of change inside the object. So this is the price we pay in a sense for quantization for this wave function stuff.

[00:26:47] Al Scott: You said something completely different can happen that you’re not doing to the particle. Could you give an example of what could?

[00:26:54] Holger Hofmann: Well, it’s the rotation angle in our case, right? So this is this confusion. I [00:27:00] mean, we think we do to the particle either the alpha rotation or 0. And then after the interference, we look at the particle and we actually, by this compensation confirmed that it was a 2/3 alpha rotation. That then leads to the conclusion that the best way to understand is that the particle was 2/3 present in the path.

[00:27:25] This also shows that there is a bit of a problem with a typical Feynman path interpretation because the Feynman path interpretation gives you the feeling that it’s either here or there even so it’s superposition, but the superposition itself creates this fractional presence.

[00:27:46] What is important, what is very confusing is, I mean, the initial wave function itself is not enough, right? If you have the initial wave function, it could still be completely random. It could be zero or one, or it could be actually [00:28:00] something else. In order to pin down the values, you need the final detection as well. So you need the initial preparation, the final detection and the combination of two coherences; creates these fractional presence, which creates this problem that you always imagine that if you want to find it in the path, you have to detect it in the path. And then it is either 0 or 1. But if you don’t detect the path, but detect something else… and there was also a development that happened in the early two thousands, right? That is in the paper also, you might find, we talk about Osawa the Osawa uncertainties and specifically the Osawa Hall uncertainties. And that has something to do with a measurement theory, introduced by professor ova here in Japan Nagoya this measurement theory evaluates the ever of a measurement when you are actually measuring something completely different, possibly.

[00:28:59] [00:29:00] So the measurement is all general. You don’t have to measure this physical property, and then you still attach a physical value to the physical property for this outcome, and professer Osawa showed that mathematically, you can calculate an error of that.

[00:29:19] And it turns out that this theory basically predicts that the weak values have the lowest error that was then discovered by Hall in Australia. So there’s a connection here with the weak values and professor Osawa is also using the initial state and the final measurement outcome in order to talk about a completely different physical property.

[00:29:41] So to understand that better, it’s good to use the idea of statistical correlations. You have initial information and then you measure something that is correlated with what you want to know. The initial information is the correlation it says for this state, the relation [00:30:00] between this property and the other property I want to know is given. So after I see the other property, I can add the two pieces of information. And it looks very much as if that was the case. And in fact, it’s very consistent only that the values you get are not the eigenvalues in this case.

[00:30:19] Al Scott: Let me go over this one more time just so I make sure I understand.

[00:30:22] So you’re firing the neutrons into two paths or into a path, and then it splits into two paths. You’re doing a, a small spin rotation in one of the paths you’re putting a magnetic field that, so when the neutrons come in, are they spin aligned? Are they, are they all.

[00:30:39] Hartmut Lemmel: Yes, they are polarized.

[00:30:41] They are prepared in a certain initial direction.

[00:30:45] Al Scott: So you give one path has a slight rotation on the spin vector. They come back together and interfere and they choose a path and then they are detected. And then you have a, a second magnetic field that can correct [00:31:00] for any spin misalignment of the particles.

[00:31:04] So you can’t actually measure how much spin they’ve had if they’re interfering. But if you rotate them back a fraction of the amount of spin rotation that was given to the one path you, you correct? All of them back to, to aligned.

[00:31:23] Holger Hofmann: Yeah.

[00:31:25] Al Scott: Wow. So you’re actually not measuring the rotation at all. You’re just canceling what rotation was given. And it’s odd to me that it was, it’s not one half. Why is it not one half?

[00:31:39] Holger Hofmann: Because we deliberately changed the balance between the path. And that has something to do with correlations. As I said before, now you can get one half there’s a slight problem with this. And actually the slight problem is not so slight because when you balance the interferometer, you have almost [00:32:00] 100% of the output only in one direction. And that creates a bit of a problem because you should be able to also see the other direction, but when the probability is close to zero it’s hard to deal with that.

[00:32:15] So we actually don’t want to have perfect interference. It’s much easier if you change the balance between the paths. In our case, we had 90% in one direction, 10% in the other. So then we can in principle, see both outputs, right? It’s not a foreseeable percent problem. But it also means something else about the changes from one half to another value, because the imbalance between the path corresponds to a correlation between the interference and the path.

[00:32:47] So the correlation is then changed; the one half means basically no correlation. So you measure the outcome and all you get is a [00:33:00] confirmation because in the one half case, you would initially know it’s equal probability in both and therefore equal average intensity, both. And it just confirmed that this is always the case and the particle really splits one half, one half. But our case whether I measure the 90% port or the 10% port really makes a difference because there’s a correlation between the outcome and the path. So in the 90% outcome port, the constructive interference, we get two thirds, but in this interference at 10%, we get two; looks as if there’s a double neutron in the one path.

[00:33:44] Al Scott: Okay. You’re throwing a wrench into my thinking here.

[00:33:48] [Laughter]

[00:33:49] Holger Hofmann: Well, yeah, it is not that simple because these are the anomalized weak values that have confused a lot of people.

[00:33:56] Hartmut Lemmel: By the way Holger, I mean, this ratio we have chosen is [00:34:00] arbitrary. We can prepare it. Yes. To have any ratio. We, chose some nice values. So that’s why it’s one third and two thirds. But actually these values are too nice because two people already asked me, well, it’s probably related to the quark structure. You know, every neutron has three quarks in it. So one goes left. Two goes, right. I just want to clarify, this is not the case. The quarks are not separated. We don’t care about the internal structure of the neutron it’s the whole neutron which goes to one third here. And two thirds there, for example.

[00:34:34] Holger Hofmann: Yes.

[00:34:34] And it should really be said that the one half, one half case also exists, but it causes problems because of the zero probability prediction for the significance. If you are doing the experiment too well, you are in real trouble. That’s not what we wanted to have.

[00:34:52] Al Scott: Wow. Okay. So, but this is cool to me, the philosophy of the interpretation of quantum mechanics, the, how this [00:35:00] experiment bears on that is really interesting. Cause I never thought anyone would be able to come up with an experiment that would test say the many worlds theorem or the pilot wave.

[00:35:11] But we have these theorems that kind of, posit a real particle that people, you know, people are very uncomfortable with having wave functions represent the root of reality and that, you know, they want to have real particles existing, but quantum mechanics tells us that there’s no hidden variables, that, you know, particles when they’re not being observed, don’t have paths. And so we’ve come up with these philosophical ideas to maybe explain how they could have paths and maybe the, you know, many worlds theorem, you know, posits that the wave function never collapses and that an observation is just telling you which universe you’re in, and each particle does exist and, and has real paths in all of these different universes, and you just don’t know which universe you’re in until you’re measured. I think this [00:36:00] makes that an untenable interpretation…

[00:36:03] Holger Hofmann: Well, it’s certainly a critical commentary on this. As I said before, for me, it’s for a long time, it was really essential that there needs to be something you can observe.

[00:36:16] And the weakness of the theories that you just listed here and described very nicely is always that these things are not observed. And you are trying to somehow fit your idea of reality, which is not based on observation into the observable universe. For me, I’m always wondering how can I have confidence about such a thing when my only confidence about reality in everyday life is the observation?

[00:36:48] And I think that’s my criticism of the theory. So what I think we have done, and which is great is we have actually demonstrated that you can be a bit more careful with the [00:37:00] experimental investigation of questions, like where the particle is, and it is possible to design in this case, not really a measurement, but a verification method that tells you something about the physics of the presence of a particle, which is not quite the same as the reality of a presence of a particle. And, and that’s maybe what course is the confusion, because the idea of reality is going a little bit beyond effects and observations. And that’s, I think there’s a misunderstanding that the people who think they’re realists think they’re not philosophical, but what you mean by reality is a very philosophical conversation.

[00:37:50] Al Scott: It gets back to the, you know, the original controversy over dualism, right? You know, the particle is both a wave; it’s both a wave and a particle, depending on [00:38:00] what it’s doing. And, and if you’re not observing it, it doesn’t have a reality. It’s the dreams that stuff is made of.

[00:38:08] Holger Hofmann: Yeah, you could say that.

[00:38:11] Al Scott: How certain are these results?

[00:38:12] What are possible possible confounding factors? Is there anything that causes you to hesitate about this result?

[00:38:20] Holger Hofmann: Well, for my side, I’m pretty much aware of what some of the criticisms are because I get them all the time. I have been getting them for years and years and I take them seriously. So I’m not dismissing that.

[00:38:34] So what the criticism here is basically is in this idea that you know where the effect comes from. So it is actually a, a bit of a problem that we also discussed at length that we are now saying that this spin rotation definitely comes from the presence of the particle, but there is of course, a school of thought that [00:39:00] tries to give more reality than that to let’s say the mathematical description.

[00:39:10] So there is an idea that the interpretation that this rotation originates from, the presence of the particle or from the magnetic field that the particle saw, has the flaw. But the way we calculate this does not include a mathematical definition of what a force is, a mathematical definition what a magnetic field, what the quantity of the magnetic field really means, let’s say that. So there is this problem of the mathematics. And that’s a huge impact on, for example, many world interpretation, which basically says that; it says the mathematics is in some sense superior to arguments like ours, where we are saying that [00:40:00] for cause and effect reasons it should be the presence of the particle that makes the rotation two thirds. And it can be very difficult to argue with these positions because in some sense, these are Platonist ideas of physics. And in Platonism or in a trivial version of Platonism, the idea is always that we are just in this cave and see shadows on the wall. So the experimental side is a mere shadow of a higher reality.

[00:40:33] And so it’s kind of then saying, well, the mathematics is so beautiful. Why are you dragging it down to this rather low level, right. Of actually taking this compensation so seriously and that also has to do with then the question, have we understood because of this experiment, one could actually argue that we have a new idea what entanglement is. [00:41:00] But then a lot of people would actually now say entanglement is a thing in itself like energy or so it’s a resource. So we are happy to basically say there are no individual objects. Anyway, everything is entangled and we are cutting that deliberately into pieces again. That is why the first argument in the first discussion has to be, is it possible in modern physics and modern quantum mechanics to agree that we have to put experimental evidence first? Basically the argument is skepticism. You say you have experimental evidence, of course, the click in the detector and so on that is not controversial, but you shouldn’t argue that it shows anything. The mathematics tells you what the experiment shows. And you see that a lot, like for example, many world theory and also Bohmian mechanics.

[00:41:57] That’s what I said about the Bohmian mechanics. The clicking [00:42:00] the detector is taken as proof for the whole existence of the particle before, but it’s just this end click of the detector. And then the argument is that it’s done for the beauty and completeness of the theory. And then people normally argue in the opposite directions is we want to believe this theory and now we prove it somehow.

[00:42:24] And I’m just thinking we should. Only talk about experimental evidence and try to find a large variety of possible experiments.

[00:42:36] Al Scott: This is how we learn about what is the, the proper interpretation. I think this, we need more experiments like this to inform a better understanding of, of the, of the physics, because really quantum mechanics is an unfinished theory.

[00:42:50] It doesn’t describe the measurement problem. It doesn’t, you know, there’s the Copenhagen interpretation that the wave function is a probability function and [00:43:00] shut up and calculate.

[00:43:02] Holger Hofmann: I would have, I would actually say I saw you description of the Copenhagen interpretation. I would modify that a little bit because I wouldn’t say I grew up with that, but almost grew up with the teachings of Heisenberg, especially. And there is a lot of depth in the Copenhagen interpretation. The biggest problem is it’s a kind of agnostic interpretation. You are right about the shut up and calculate in German I would call it the denkverbodeIt’s not allowed to think. But otherwise it’s very deep because it basically emphasizes that you cannot really know about things that cannot be observed. But the big problem is that Copenhagen also doesn’t actually tell you how to judge what you can observe that’s where the problem is, right? Yes.

[00:43:52] Al Scott: It doesn’t tell you what an observer is or what classifies an observation. Yeah.

[00:43:58] Holger Hofmann: And also what things [00:44:00] you can actually observe and whether there are things you cannot observe.

[00:44:04] When I was studying, I was, there was, I remember one time where I was really a bit upset from the quantum mechanics lecture, because everything was about the Hamiltonian and energy. And I think what does it mean? Can we only measure energy suddenly? I mean, that doesn’t make sense because actually energy measurement is rather difficult. You don’t have any intuition of an energy measurement really, but still, I mean, they’re taught you have to diagnose the Hamiltonian and nobody really explains why.

[00:44:33] Al Scott: I think it gets back to conservation laws, right? I mean, you have these conservation laws and these things, you feel must have some reality to them because they’re conserved. Yeah.

[00:44:43] Holger Hofmann: Thermodynamics of course is very important. So thermodynamics uses the energy and thermodynamics uses little else.

[00:44:50] We did mostly solid state physics and solid state physics you start basically with a material that has a temperature, and then you perturb it a little bit and [00:45:00] then something happens. And that is most easy to describe using the heightened states of energy. But that’s the problem that you cannot really easily generalize that. So when I, I learned about optics, this whole thing collapsed very quickly because in optics, that’s absolutely not the case. So laser light is a coherent superposition of different photon numbers. It’s absolutely necessary to use that because you get macroscopic waves.

[00:45:32] Al Scott: Interesting. Yeah. And so what is your interpretation of quantum mechanics? What do you think? I mean, this is something that we all want is, you know, a better understanding of reality and, you know, is there reality to understand even, I dunno what to think.

[00:45:49] Holger Hofmann: Well I wouldn’t call it an interpretation really, but I have a tendency. We basically, what I’m looking for in quantum mechanics is an [00:46:00] explanation of actual experience; the way we experience the world should be explained in a consistent manner. And one thing where I’m a bit different from a lot of colleagues at least, is that I’m basically saying for me, reality starts from our side, from what we experience. And then we have to try to find our way all the way down. And the only way to do that is to understand the relation of causality that allows us to build all the machinery, like in the neutron interferometer, right? So there’s a causal connection here. We built this, these objects, we make our crystals, we put them in the right place, we make sure they don’t shake and so on. And these are all causal chains. They’re basically creating the causes that ultimately will create the detection of the neutron at the very end. It’s a long [00:47:00] chain, but in this chain we rely on causality. We rely on laws of physics that are not about reality at all. They are just about relations.

[00:47:12] I mean, for example we do say, I mean, Hartmut is setting up this crystal and we talked about, this is amazing that it’s basically precise down to the atoms. But of course that is not, you don’t basically see the atom when you do that. So the interesting thing is that you can do that even. So you’re handling this big block of material, you’ve just made sure it’s very good, very pure material.

[00:47:38] And then I don’t know exactly what message you used to fix that I was always thinking about this, you know, Bohr had all of these old pictures with the slits that are screwed down and I sometimes saying to students, you know, the, the thing is when you screw down something that’s elastic, so you’re basically fixing things with a kind of [00:48:00] spring. Yeah. So it’s obvious that even though you don’t see it, everything is shaking. And it’s interesting that I’m, we are not normally aware of that. We just ignore that. And then we have to use some tricks if the experiment gets so difficult that you have to be very precise, right.

[00:48:21] Al Scott: So you’re basically positing that we’re not attacking any underlying reality here that we’re measuring relations between observables and, you know, you can be agnostic, you know, as it, what was that German word? What do you call it? Don’t think don’t go beyond the, the math.

[00:48:46] Holger Hofmann: Denkverbode is a negative word. I wouldn’t endorse the Dekverbode, quite the opposite. And also I would say here I would say for me atoms are real enough. I had this there was an interesting conversation with EF Steinberg years back when he actually visited us in Japan. I mean, he, he did all the weak [00:49:00] measurement experiments and to Toronto University. So we had an interesting conversation about the question of realism, right?

[00:49:07] So he was listening to me and then saying, but you know, you sound as if you wouldn’t believe that there’s a backside to the moon because you haven’t seen it. And I said, no, no, no. That is a complete misunderstanding. Of course, I very much believe there’s a backside of the moon because I know that I could see it if I went there and that’s quite enough.

[00:49:30] So I’m not the fanatic kind of here, but I would want to know how do I experience that? So, yeah, atoms are very real because we have several ways of experiencing atoms up to the point of putting one atom in a trap and actually seeing it.

[00:49:50] I think we should take that more seriously. That objects become real by interacting and most objects interact enough.

[00:49:59] Hartmut Lemmel: But still, [00:50:00] you can build an interferometer using atoms, even though you can use large molecules and build an interferometer.

[00:50:08] Al Scott: So it depends on the context, whether you can see them, if they’re in this special interferometer, that Hartmut has built suddenly they’re not real.

[00:50:15] Yes, yes, exactly.

[00:50:18] Hartmut Lemmel: They are still real, but they can be simultaneously at two locations.

[00:50:22] Al Scott: Interesting. Wow. This is great.

[00:50:25] Holger Hofmann: Yeah, but that depends on the interactions and this is the thing, the microscopic objects interact massively, and that makes them very robust. Right. They’re always telling the students. And actually I did a calculation when I was first time in Japan as ’95, actually I wrote down a calculation quantizing the moon, cuz as you know, Einstein raised, this question is the moon there where nobody is looking. Now my, my answer to Einstein would be spontaneously. It’s pretty hard not to look with regard to the moon. The moon [00:51:00] is hard not to see and, but there’s actually a calculation here and it’s a fun calculation. I mean, I hope I get the numbers roughly right.

[00:51:09] So the angular momentum of the moon around earth is 10 to the 68 H bar. And when one photon comes from the sun and is scattered to us from the moon that changes the angular momentum of the moon by 10 to the 15 H bar. One photon. And you never really know how many photons are scattered by the moon because it’s statistics. And basically the root N photon is noise. So basically the moon is experiencing quite a diffusion of the angular momentum all the time.

[00:51:48] That is the measurement back-action here, really. So it’s no wonder that the moon is easily localized. It’s an open system it’s super noisy. So this argument [00:52:00] is very easy to make. And I would be saying just that the moon is really the sum of all the effects that the moon has, only the moon has lots of effect.

[00:52:10] Al Scott: Right, right. It’s very well tied to the universe through interactions. That’s an interesting way to think of things.

[00:52:22] Holger Hofmann: It’s basically screaming loudly, “Here I am”.

[00:52:24] Al Scott: So this is a ground breaking paper. I’m sure you’re gonna get a lot of publicity from it. What, what are you guys thinking about working on next?

[00:52:31] Hartmut Lemmel: Well, I have thought of, I mean, we have done different experiments also with weak values. Like the quantum Cheshire had different things and, but they were measured always with, as a weak measurement, which applies only to the whole long solver. Someone could think about which experiment can be refined by the new method to get really answers about each individual particle.

[00:52:57] Then what I also find [00:53:00] interesting right now, we have used this, the interaction between spin and magnetic field. What if you, if we can do something with the neutron, mass and gravity, for example, is what this be possible. So look, our new neutron is distributed in the gravitational field for example, but that’s a much weaker, so it’s, it will be difficult.

[00:53:27] Holger Hofmann: One thing I would maybe want to suggest sometime is also to look a little bit at phase shifts. It could be done in something like a Cheshire cat such scenario too. But there’s an interesting thing about the uncertainty of our method. So there is actually a natural that immediately occurs when you look at phase sensitivity.

[00:53:49] And there’s an interesting relation to the information about phase that you can get out of an interference experiment and the uncertainty of the path. So we have been [00:54:00] operating in the regime where you get no real phase shift information where the interference is super stable against small phase shifts. If you go away from that and go on a where it’s sensitive to face shifts, then actually you do observe fluctuation. Then you start to observe statistics again. And of course, I should also say here I’m just getting a bit busy with two doctoral students who are trying to work in the theory direction on the topic. And I think it’s worthwhile to really try to find out every detail of quantum interactions.

[00:54:39] So the thing that we really don’t understand enough is how quantum systems interact and communicate the information about their properties. Once it’s clear that the reality of the object comes from the interaction, it requires a much more detailed and much more [00:55:00] resolved look at how interactions really work. I do hope that the young guys do a little bit in that direction because just they should be getting in on it too.

[00:55:13] Al Scott: Oh, very interesting. Well, I think we’re getting to the end of our time slot here and I appreciate you both for coming on and chatting with us about this amazing experiment and looking forward to, to more in the future to elicit some more information about quantum mechanics and reality. And for coming on the show, I’m gonna send you guys both a t-shirt for The Rational View. So thank you.

[00:55:34] Holger Hofmann: Thank you.

[00:55:37] Hartmut Lemmel: Thank you.

[00:55:42] Al Scott: If you’d like to follow up with more in depth discussions, please come find us on Facebook @ The Rational View, and join our discussion group. If you like what you’re hearing, please consider visiting my patron page: @patron.podbean.com/therationalview. Thanks for [00:56:00] listening.

Leave a comment

Your email address will not be published. Required fields are marked *