– What do you think is the strongest argument today that gravity really does need to ultimately be quantum mechanical?

– Yeah, but if you put a mass in a superposition of two different places, you are really faced with a very stark choice there if you do not allow gravity to also be quantum mechanical and to respond to the mass in each of these places simultaneously.

– Welcome, everyone. Today’s conversation is in the arena of quantum mechanics, gravity, quantum information, trying to understand, you know, the deep structure of reality. And I’m pleased that the conversation today is with Vlatko Vedral, who’s a Professor of Quantum Information Science at Oxford University, my alma mater, I might say. And I’m so pleased that we have this time to talk about these ideas.

– Thank you for having me.

– Also, you’ve written a number of popular books that have brought these ideas to a wider audience, and so these are things that I think really captures people’s attention. So I thought we’d begin with the following historical trajectory, which I find deeply inspiring, which is there were questions about quantum mechanics in the 1930s raised by Einstein and collaborators having to do with a quality of the world called quantum entanglement that we’ll have a chance to talk about. But the reason I raise it is at that time, it wasn’t really something that experiment could really weigh in on. And yet, if you fast forward 40, 50, 60 years, or even close to 100 years, not only has this come within the realm of observation experiment, we now have harnessed it. And we use this idea of quantum entanglement and building things called quantum computers. So, do you see that paradigm as, say, playing out again in the kinds of work that, for instance, I’ve spent time thinking about you, as well, how you put gravity and quantum mechanics together?

– It’s a great question, actually, and I think the analogy is really perfect to compare these two. You’re absolutely right that there was no way to really test entanglement properly in the late ’20s and early ’30s. And probably that’s why we ended up with so many different interpretations and different views of quantum mechanics. In fact, it was even credible to a high degree to suspect that maybe it will not apply to certain systems outside of the atomic physics domain or nuclear physics. And the rate of progress in that direction, of course, was phenomenal. But I think you can see something very similar in quantum computing and quantum technologies, which I would say, you know, early ’90s, mid ’90s, which is when I started my research and my PhD, people were still skeptical about quantum computers. Even some people who later received Nobel Prizes, by the way, for their work were actually writing articles saying, “This is never gonna work. “It’s far too complicated. “We will never be able to control noise in these systems.” And then round about the year 2000, when I think researchers from the solid state domain, so superconductors, Josephson junctions, things like that, when they started to get engaged and they demonstrated that you can actually etangle two of these Josephson junctions as qubits, it changed lots of people’s minds, actually. And I think the heavy industry then became more interested, as well. And the rest is where we are now with these technologies. But I think, coming back to your question, and I think that’s the one I find really interesting is that, because ultimately I’m a physicist, I don’t really, I don’t mind if we don’t get a full-fledged quantum computer for a while. But what really interests me is these deep fundamental questions in quantum physics, they interface with gravity. And it seems to me that these technologies now allowing us to probe that, and that’s really the exciting thing.

– Now there’s an interesting article, I don’t know if you saw it, there’s no reason why you would have, in the “New York Review of Books” from, I don’t know, the 1990s, by Freeman Dyson.

– [Vlatko] Oh, yes.

– And Freeman, in that article, was suggesting that maybe gravity is not quantum mechanical. Maybe quantum mechanics really does apply when things are small. Gravity applies when things are big. And that’s the way the world is put together. And you string theorists and loop quantum gravity people and dynamical triangulation folks who are trying to somehow put them all together, it’s a misguided program. Now, I looked at that as a fledgling string theorist and said, I can’t imagine that’s how the world is. Number one, there are black holes where things are small and dense. There’s a Big Bang, a lot of matter, very small. You seem to need gravity and quantum mechanics. Those aren’t slam dunk proofs, but it really does suggest that you do need to blend them together. What do you think is the strongest argument today that gravity really does need to ultimately be quantum mechanical?

– Yeah, that’s a great point actually, that somehow, specifically because of this lack of experimental evidence, you could actually, you could credibly say that maybe the two domains will never intersect. Or maybe they do intersect, but we will never have access to experimentally probing early universe to that degree of accuracy.

– Or the inside of a black hole. It’s a little dangerous.

– Inside of a black hole. That’s it. Very, very dangerous. Exactly.

– Right.

– Place to be. But I think for me, the would be very similar to how people argued about the need to quantize the electromagnetic field, that takes us back again to late ’20s and early ’30s. And I think it’s simply the consistency, that if you have a bonafide quantum system, like you know, that an atom or an electron is in two places at the same time, you’ve really put it in a superposition of two places, it now generates the, you know, an electromagnetic feel of a certain kind. And the question is, how do I describe this now? If really the election is fully quantum mechanical, but you don’t allow the field to be quantum mechanical, it seems that there is a fundamental inconsistency there. And I think Heisenberg explored this quite a lot in the early days, to argue that actually if you really could couple a classical system like that to a genuine quantum mechanical, you’d be violating his uncertainty principle. Because basically you could now squeeze this quantum information into something that definitely is in one state or another. Which would obviously contradict the basic laws even of the quantum mechanical system that you originally assumed was quantum mechanical. So it seems to me, if you take a massive object, and we can argue how massive this object needs to be to make a significant effect gravitationally, but if you put a mass in a superposition of two different places, then you’re really faced, and I think Feynman said this a long time ago, you’re really faced with a very stark choice there if you do not allow gravity to also be quantum mechanical and to respond to the mass in each of these places simultaneously.

– Right. Now, it’s interesting, because Heisenberg and Bohr, at least before issues of politics in World War II, you know, got into their relationship, had a very tightly in sync perspective on quantum mechanics. And yet, Bohr proselytized quite effectively for his era that, you know, when things are big, they’re just classical. And that’s what you just have to deal with. Was he closing his eyes? Was it wishful thinking? Was he not thinking as clearly as he should have? Or like, is there a chance that he’s actually correct?

– All of these are possible. Exactly. As you said, all of these options are possible. It seems to me that there was, at that time, I don’t know where that comes from. Of course, you can justify it intuitively because we all experience some kind of classical world around us.

– Yeah.

– And all our communications, all our information is somehow classically encoded. We always use distinguishable, fully distinguishable, there’s never this quantum ambiguity of a superposition in our communications. I think where they make a mistake, I would say, people who subscribe to this logic, is that they then conclude that you must have a classical world outside, rather than just being happy to live with an approximately classical reality. So I think even if you believe in the fully quantum universe, we will all acknowledge that quantum effects diminish frequently with the size and the complexity of the system. Simply, there is so much noise that all of these entanglements and superpositions get washed away–

– Washed out.

– That’s it, washed out. That’s simply the explanation I would say. So that’s not in contradiction with the system being fully quantum mechanical. But I think to say, absolutely you must need a, you know, you have to have classical operators, classical observers, all of these things that they insisted on, it seems to me that that’s not necessary. But as you said, we haven’t quite tested nature at that level yet. So it is possible that they’re right.

– Right. I’d love to get into some of the tests that you’ve been thinking about. But before we get there, there have been approaches, mathematical approaches that have tried to, I don’t know if the motivation was to realize Neils Bohr’s vision, but for instance, the so-called GRW approach to quantum mechanics, where you take, as you know, Schrodinger’s equation, you add a new equation that actually has the effect that when you have a small number of particles, it has almost no influence. But if there are many particles, it does, in effect, what Bohr wanted. It gets rid of the quantum uncertainty, it gets rid of the superpositions, and you are forced, through this new mathematical term, into a single definite classical reality. Do those so-called spontaneous collapse theories, do they work for you? Do they have any draw on your attention?

– No, I think I investigated them. But more thinking how one would put a limit on these rates of collapse. Because obviously if you put the bound that’s high enough, then all of the current experiments that have verified superpositions will anyway pass that. There is no conflict. To me, it’s problematic more conceptually, right? That the problems are then that again, you are forcing something to really become classical by introducing this extra term, as you said, into the, into the Schrodinger equation instead of just saying, well, that can happen spontaneously if your system interacts fully quantum mechanically with another system, what we call decoherence.

– Decoherence. Right.

– Decoherence would explain the same thing. And I think where this really doesn’t work, this kind of hybrid model, half classical, half quantum description is if you, for instance, think that our symmetries, our conservation principles, let’s take conservation, if you believe that energy is conserved in the absolute sense of the word conserved, rather than stochastically conserved.

– Average.

– Average. On average. And I think all of these collapse models have that problem, that that half of the times you will get more energy, the other half you will get less energy. Whereas everything we’ve done so far suggests to us that energy is conserved in each branch of these, identically conserved. Momentum, as well. Anything else. So, what would bother me is simply that suddenly all of these very fundamental laws and principles or what we thought were fundamental are becoming now stochastic.

– Right, right.

– And it just looks ugly somehow.

– No, no, I share a similar view. Of course, aesthetics takes you just so far in revealing truth. But certainly it guides our intuition. But one thing that might be worth spending a moment on, so you mentioned decoherence, and it’s not a topic we’ve spent a lot of time talking about in these conversations, ’cause it’s a little bit hard to fully explain. But roughly, it’s the idea that if a system interacts with an environment that can impinge on it in many different ways, that tends to suppress the quantum-like qualities of the system and make it appear more and more like a classical system. But one thing that it doesn’t do is it still allows for many possible outcomes. It doesn’t pick one of the many. Whereas the GRW approach does pick one of the many. Now, if I’ve done my research, I believe you come at the world, I shouldn’t say the world, you come at quantum mechanics from a somewhat many worlds perspective.

– Yes.

– So presumably it doesn’t bother you that decoherence doesn’t pick out one outcome and allows many worlds.

– That’s right.

– Whereas for someone like me, and I’d love by the end of this conversation, you convinced me I should be a many worlder, you know, that’d be beautiful. But for me, that is one thing that the GRW approach does give.

– I understand. It does give, and and I think that’s probably the main motivation, that somehow, almost goes back to what Einstein complained about. You know, I don’t believe that, you know, God plays dice with the universe, whereas we have to fully acknowledge that even if you observe a single quantum bit, you know, you have a binary choice in a superposition. And basically you’re saying, now I’m gonna look at the system. Where is it? Is it on the left hand side or the right? Even then, we basically have to acknowledge that this is completely random. There’s no way to predict the outcome in this case. And so I think the GRW intuition tries to go beyond that and to say, well, maybe there is something that really chooses one of these two alternatives. Again, my problem with that would be to very similar to again what I was saying, how Heisenberg was arguing about quantizations of fields simply because charges originate these fields. And therefore somehow they have to be behave in sync. There has to be some common description of these two. So I would say if you think that the apparatus which you use to make this observation, including ultimately your own brain, your own perception, if that’s where the whole thing stops, if you really think that that’s described quantum mechanically, or as, as well as it’s described at present, you know, there may be another theory later, but at present we have nothing to really suggest otherwise, then somehow forcing deterministically a single outcome really is tantamount to violating the uncertainty principle. And so that’s what would force someone like me to acknowledge that all of these outcomes do actually simultaneously exist, because it’s the only way to preserve the consistency of quantum mechanics. I should say something about the language there. I don’t quite like the the many worlds because I think–

– You mean that name?

– The name.

– Okay.

– I don’t like it. So I’m kind of even more extreme in the sense of quantum mechanics.

– You just want us to talk about Q numbers or?

– That’s right, I think Q numbers in the sense that I think many worlds is still just one special extreme case that you can get in the universe, and it’s still a superposition of classical realities. Whereas actually quantum mechanics allows all of the states in between, it’s like exponentially.

– So it’s many, many, many, many worlds.

– That’s it.

– Where world now has a much broader definition.

– Indeed, indeed. So I think that’s the way, and that’s what gets you into trouble with these hybrid descriptions. I think if you acknowledge that the world is like this, then it’s very difficult to preserve any consistency.

– Right. Now we are sitting here in a room in Oxford. Down the road is the Oxford Mathematical Institute where I actually began my graduate studies for just a short period of time under Roger Penrose. And I think many in our audience who follow these kinds of developments are at least familiar that Roger Penrose has put forward his own approach to some of these issues where gravity and quantum mechanics do have a conversation of a very particular sort that he thinks could resolve some of this issue of picking out one reality or another. Has that approach carried the day at all for you or?

– Again, I think I would label it as one of the hybrid semi-classical approaches. It seems to me we have these two phenomenally successful theories, and somehow it is a shame that we have to use them in completely separate domains. Any physicist, I think, would prefer to have a unified description of reality. And now the choice you’re facing is, well, there’s more than just the two choices, but one choice is certainly to force gravity to behave quantum mechanically, to quantize gravity. But the other one, and it seems to me Roger Penrose is fond of that approach, is that he trusts general relativity, he trusts gravity, the theory of gravity more, and he says can we gravitize or geometrize, or whatever the word would be there, quantum mechanics. And I think it’s that intuition which ultimately tells you that if you really take general relativity, it’s a fully classical theory as far as information is concerned. Things are in definite states. They are never really in quantum superpositions. In that sense, if that’s your intuition, then it must be somehow that you ought to be able to force quantum mechanics to make these choices. And so I can fully understand where he comes from. It’s just, it seems to me that all of the experimental evidence we’ve had so far, plus the all of these consistencies that I’ve been talking about are probably going to force us to go the other way.

– Yeah, no, I of course have that same perspective. I imagine that historians or sociologists of science would look at folks like us and say, yeah, you guys came up through the ranks of quantum physics. And Roger Penrose and Stephen Hawking and others in that arena came up through the general relativistic approach. And you tend to stick to your training. And our training suggests that quantum mechanics will prevail, theirs suggests that general relativity will prevail. And so it’s sort of a natural.

– It is very natural. It’s very natural. Yeah, he takes, for instance, the equivalence principle as a classical principle. And I think if one really does that, if you really believe that this is upheld only classically, so the equivalence principle suggests that you can never discriminate gravitational acceleration from any other form of acceleration, if you’re really within that system.

– In a sufficiently small region.

– In a sufficiently small region. So Einstein, I think, had this falling lift. Or the lift that was accelerated and basically feel pressure.

– You’re glued the floor.

– Exactly. Indeed. You’re glued to the bottom. And then you’re asked the question, are you in the vicinity of a very massive object, or is someone really pushing you from behind? And we know that we can’t tell the difference there. Now, if you believe that this principle cannot function in a superposition of states, and that’s what Roger would say, Roger would say, if you have a massive object in a superposition of two different places, then a test mass, a separate test mass will have to now decide whether it goes one way or the other, whether it obeys the equivalence principle in one branch, if you like, or the other one. And you can see how that links back to the GRW logic. GRW didn’t really even supply the mechanism behind collapse. So Roger, in a way, Roger’s logic would be perfect.

– That’s right. Fostering that, yeah, yeah.

– That’s right. He would even say, here is the mechanism, it’s because of the equivalence principle. But a person like me looks at all of these other principles, and I think, well, wait a second, we’ve done exactly the same successfully with energy conservation, momentum conservation, all of the symmetries that we think the universe obeys are successfully absorbed into our standard model, if you like, the quantum description of all the other forces. So why not the equivalence principle? What exactly would prevent us from actually accelerating in two different ways simultaneously? But no one has done that experiment. And I believe that’s why.

– Right.

– It could still be that we are surprised.

– And that’s what I’d like to, you know, go down that direction. I mean, from our general agreement on how we look at gravity and quantum mechanics, you might think that we’d both be string theorists.

– Yes.

– I am. But you have taken a different tack toward trying to really sink your fingers into the conversation between quantum mechanics and gravity, even proposing experiments you think that are actually doable.

– Yes, yes.

– Now, the one thing I would quickly say, just as a prologue or preamble, you talk to a string theorist like me, and the natural question I get all the time is, all you do is mathematics. You’ve got no, and I have to sit there and turn a little red, and not really, you know, I shake my head. Yes, it’s just pure mathematics. Now, just mathematics, you know, I don’t know that you need the just, but I get the point. But the reason why it’s tough to test string theory is when we look at where string theory tends to manifest itself, it’s on extraordinarily small scales, like so-called Planck scale, 10 to minus 33 centimeters, extraordinarily massive densities, you know, or mass scales like, you know, the Planck mass, you know, 10 to 19 times the mass of proton. And we’re like, yeah, it build us a big enough accelerator, sure, we’ll test these ideas. But you suggest that at least to understand something of the conversation, not necessarily string theory, per se, of graduated quantum, we might really be able to do that.

– Yeah, I think that surprised me. You’re right. I wrote a paper with Chiara Marletto, he was my collaborator almost 10 years ago now in 2017, where both of us, and separately there was a group at University College in London, Sougato Bose and his collaborators. And what surprised all of us really independently, we all thought differently about this aspect and maybe came from different directions, but the surprise was really that some of these tests, whether gravity at least has some quantum degrees of freedom. So we are not as ambitious as what you’re saying to test the final theory. And this experiment is not able to discriminate between various canonical–

– Any of the approaches.

– Indeed. I think string theory would pass it just as well as.

– Loop quantum gravity.

– Loop quantum gravity. Anyone would agree at that point with this. However, all of the hybrid, certainly any collapse model, no matter what the origin of your collapse is, would actually fail this kind. So, any semi-classical description where gravity still stays classical would not be able to pass this. And what really surprised us is that the masses you need are actually smaller than Planck’s mass, because all of these intuitive arguments are telling us you need something at least of Planck’s mass, superpose it, and then we are talking about gravitons and so on. So here you are not detecting gravitons, which is, that’s the interesting bit as well, that you could even have quantum approaches, which don’t necessarily, like loop quantum gravity. For them, the fundamental elements are really areas and volumes that you are quantizing. And somehow gravitons are a derived concept out of all of that. And it’s a very simple experiment. I can describe it.

– Yeah, please. I’d love to hear it.

– It’s a really simple experiment. So, it really goes in the direction of what Feynman suggested. And Feynman said, take a massive object, put it in a superposition of two different places, and then simply interfere it at the end. So let it in couple to gravity, which would happen naturally. Then perform, bring these two possibilities back and see what kind of interference do you get. Do you really get quantum interference fringes? Or is it just stochastic in the classical way? Now the problem with this experiment, this would be a phenomenally difficult experiment in its own right, and certainly it’s worth doing and it will be an experiment along the way.

– And what sort of masses would you need to put in different places?

– I think for us, the masses were even smaller than a nanogram. So, certainly 10 to minus even 15, 16 kilograms will do the job. So we are talking about five, six, seven, eight orders of magnitude smaller than Planck’s math. Still extremely challenging. No one has done this kind of experiment before. But it seems to me quantum technologies are getting closer and closer. So what we need is, we can’t conclude anything from a single mass for the simple reason that a single mass would interfere or not interfere. In both cases, gravity could be quantum and it could prevent the mass from interfering by entangling itself to that mass, which is what Schrodinger taught us. Would be standard decoherence. Or gravity could be classical, and the classical noise could affect, which would be more like what Roger Penrose is suggesting. So basically, any outcome in this experiment, interference or no interference, you could equally well have a classical or a quantum model of gravity. And that kind of gave us an idea that maybe we need another mass in a superposition to test the gravitational effect of the first mass. So we are going in that direction of the kind of the equivalence principle. And so now, you double this, you have two interferometers. In each of them, you have a massive object. So we are talking about similar masses in each of these interferometers. And you now have two branches here and two branches here. And the question is, does gravity now couple each of these branches simultaneously? That’s the intuition behind.

– And so these are literally interferometers that are near each other?

– They’re near each other. Yes, so they would be, in one of these important–

– Or just in a sufficiently, a gravitational field, just different heights in the gravitational field?

– Even that is possible actually, to do it in free fall. It doesn’t, the experiment that we have in mind is to trap one of these nitrogen-vacancy diamonds. So, what you have is you have a spin of this system. It’s large enough that it’s massive enough gravitationally, but it has a single quantum degree of freedom that you can actually prepare in a superposition of states. And what you’re effectively performing is a Stern-Gerlach experiment. If the spin is up, you send a nanodiamond one way. If it’s down, you send it the other way. And you do the same with the other nanodiamond. And now you let them couple. And exactly what you asked. The key question is the one you asked, if they’re too close to one another, you want them as close as possible because of gravity. You don’t want to have to keep this superposition for too long. because of course it may decohere due to all sorts of other effects. So, the distances, if you make them too small, the problem is that other forces may come into play. So even if you take neutral systems, you may get all sorts of induced dipoles, Van der Waals forces, and so on. So this is where you have to do a serious visibility study. I think what we calculated roughly is 100 microns distance, roughly micron size superpositions of this object. And you have to keep them coherent for roughly a second. We are talking about second. Extremely challenging. All of these are extremely challenging. But if you can do that for long enough, and if you really observe entanglement generated between these two masses, so what that really means is you measure one mass in one position, then you know where the other mass is.

– Sure, yeah.

– You literally confirm quantum entanglement on these masses. Then the only explanation is that the mediator, the gravitational force that mediated that must have had some.

– Quantum quality.

– Quantum quality, that’s it.

– Yeah, I see. And so, is this just ideas or? I mean, is this?

– This is, at present just ideas, but I think there are at least five or six groups that I’m aware of who are racing to actually implementing this.

– Really?

– Indeed. So people have already, a colleague of mine in Israel, Ron Fullman, has already implemented a single Stern-Gerlach, unfortunately with a Bose condensate. And I say unfortunately because it’s not massive enough. We are talking about even the largest Bose condensate has no more than a billion atoms. So we are talking about almost 10 orders of magnitude away from the masses that I need in my own proposal. But he managed to split the Bose condensate to make it exist in two different places. In fact, to even make it accelerate differently. One branch was in free fall and the other one was stationary as a reference branch. They’re beautiful experiments. And actually, quantum mechanics works perfectly fine there, obeying simultaneously the equivalence principle in all of these.

– Oh, my. That’s wild. I mean, so that’s the kind of thing, I presume that Neils Bohr would look at and start to change his perspective.

– That’s right, that’s right. That’s the question. Because I understand the intuition that gravity of course has many features that are different to other forces. You know, universality is clearly there, that it really acts on anything.

– In the same fundamental way.

– In the same fundamental way. And that’s surprising. It’s a mystery to us. Why is it like that? So, it’s not illogical to say that if something is gonna go wrong with quantum mechanics, and of course we’re all open to that, something may be falsified by future experiments, then gravity could be a culprit, more likely than something else that==

– Yeah, it’s the only one that actually speaks to the fabric of space.

– That’s it, exactly.

– Time directly and so forth.

– Exactly.

– I mean, you know, do you have a preference? I mean, would you?

– That’s a great, that’s exactly what a theoretician like myself thinks about. Because you’re really thinking, well, okay, you are now confirming gravity in this very simple, what we call the linear regime. You are really treating gravity like electrodynamics. And even your calculations are very simple. Which is why I said all of these approaches to quantum gravity will, in that special case, agree with one another. And in fact, you are right, when I go to conferences with participants from these areas and I present these ideas, no one’s even said, we’re all betting on quantum mechanics passing this test. You know, there is no surprise there. So I’m almost, I would almost like quantum mechanics not to pass. It would be more interesting for all of us.

– It would shock us all.

– It would shock us all. So the next question, which is I think the one that you asked now, and that’s really interesting to me, and I wonder what you think about it, actually. Because the question is, is gravity really all about spacetime? Which was Einstein’s initial intuition. Or should we really think of it as just another field theory, which, you know, people like Feynman, de Witt, there was that school of thought in the late ’50s and ’60s who just proceeded in the same way that we proceed with electrodynamics. How would you even answer that question? So, are these quantum degrees of freedom that are helping us entangle these superpositions, are they really quantized units of space and time? Which I think is a hugely exciting question. Or are they really just, are we still gonna remain with a kind of classical background spacetime? Which we must use for some kind of convenience, but it’s not really, it’s a fiction, it’s not the fundamental entity. And is it really the fields?

– Yeah, I think most of us agree it has to be the former, not the latter. Especially, as now, there have been these wonderful calculations, of course, and string theory, and always need to emphasize that. ’cause people sometimes get the wrong impression when we talk with such enthusiasm that the thing we’re talking about is actually tangible and real. But there are these calculations where we can study regimes where spacetime has not really yet formed, but rather we’re talking about the kind of pointillist approach to spacetime. It’s been pulverized into more fundamental entities. And it all comes from quantum entanglement. Basically, it seems that quantum entanglement, which perhaps I should quickly say is just, you know, you can have in basic quantum mechanics, you have two particles that are far away from each other that sort of have an invisible, I don’t know, quantum thread, quantum connection, that correlates their behaviors even over large distances. It makes it appear as though they’re right next to each other, and yet they’re far apart. And so it appears in certain string theory calculations that threads of quantum entanglement actually stitch the fabric of space. And if you can mathematically snip those threads of quantum entanglement, then the fabric of space falls apart into more fundamental ingredients that we’re still struggling to fully articulate what they are. But it’s clear that there are realms in which Einsteinian space and time is not there. And yet, quantum mechanics seems to persist. And so that would suggest that the underlying ideas are those of quantum mechanics as opposed to the underlying ideas being those of Einstein.

– Yes. I like that very, it really resonates with me that somehow it’s Q numbers ultimately. It’s a very abstract form of reality, and I appreciate, that’s why I think many people, even including physicists, complain about that. Could this really be the ultimate reality? But it seems to me we are stuck with these entities that are not really describable by real numbers, right? You have to assume this more dimensional quantity. And then they don’t commute. You cannot measure them simultaneously. You have to be careful how you manipulate them and the order in which they do. But I think this would resonate with me, that reality consists of these Q numbers.

– Now, one of the things though that one can raise in trying to interrogate a reality described by Q numbers, and perhaps I should again quickly say, Q numbers, what are C numbers, classical numbers? Those are the ordinary numbers that we measure. 2.7, 3.6. Q numbers are a quantum version of those numbers which embrace all possible outcomes of a given variable in a sequence of quantum measurements, like the position of electrons. It could be here or here or here or there. Collect them all into a Q number. One of the issues that’s raised, and I get your point about many worlds being a somewhat of a misnomer, but if you don’t mind as a shorthand.

– No, of course. Yeah.

– I’d like to use it. You know, people say, and I’ve had this conversation with David Deutsche, and being at Oxford, it’s a natural conversation, how do you get the probabilistic approach to quantum mechanics, which is what we have used to experimentally adjudicate that this is the right way of doing things. You measure the electron 100 times in identically prepared experiments, and goodness gracious, you find that 33 1/3% of the time, you know, just says Schrodinger’s equation, or 16% of the time here. So we’re using probability and statistics to confirm these ideas. But if it’s Q numbers realized in a many worlds approach, every outcome is guaranteed to happen with 100% certainty.

– Yes, Yes.

– How do you address the tension? Or do you see a tension?

– It’s a difficult, yeah, definitely there is a tension. I think you’re absolutely right. So, basically, you know, in the Schrodinger cat experiment, right? Where you have this quantum bit that somehow couples to the poison and couples to the cat and makes it dead in one branch and alive in another branch, you can actually, as you say, change the amplitudes for this process. You can make one branch of this extremely unlikely and the other one extremely likely. However, we would conclude that if both of them are equally real, then even in the branch with the low amplitude, the cat feels alive as much as if the amplitude was much–

– Even if it one in a billion chance. 99.999%.

– Even if it was one in a billion. Absolutely. So that’s kind of what this would conclude. And it is a weird aspect. I think we haven’t got, various people have tried, David Deutsche included, various people have tried to argue that other postulates of quantum mechanics. This is almost like the Euclid’s fifth postulate. You know, you take the four postulates of quantum mechanics and you claim that the fifth one, in this case, the Born rule emerges, but I don’t think anyone has done that really successfully. I should say, David has a very interesting set of papers where he adds a little bit of kind of postulates from betting, from economics, rational decision making. And he says, look, if I feed this into my machine, it’s gonna come up with the same Born rule. And I think that’s as best as we can do so far.

– Does that satisfy you? I mean.

– Not quite. I think I’d like to go, if you really claim that probabilities come out of amplitudes and there is nothing else there, then I’d like to see a more minimal minimalistic derivation. I think there’s still far too many extra assumptions. Other people incidentally, who are also many worlders, Liev Veitmann comes to mind, he would say, “No, you can never get rid of this.” So he even, he acknowledges I’m a many worlder, but I take the Born postulate as given, and I don’t try to derive it. It’s impossible to derive. So there are these views, as well, present.

– Right. Now, if I recall, Liev’s approach is to do with a self-locating uncertainty. So he imagines that the experimenter is in the laboratory, sets up an experiment, calculates the problem, closes their eyes, lets the experiment run, and then opens their eyes and says, “What’s the likelihood that this center of attention, “will see this, that, or the other outcome?” That seems like a, at first, I have to say, when I read that years ago, I was like, yeah, this makes sense. Like, there are many of me, but there’s only one that I’m directly aware of. And I’m asking myself, how likely is it that I’m that one?

– Yes.

– But then when you think about it a little bit more, you feel like, wait, this seems to happen too late in the story because the probabilities are something I talk about before I close my eyes. And yet somehow I can only make sense of them with this procedure, it seems, after the fact.

– You’re right.

– So the amazing thing, you’re right, I think that that also would make me feel uneasy because it also suggests that you really need some kind of almost conscious self-aware observer. And I think none of us would like that. I really do not believe that observers should have a special status in our theory. I think theyre just another collection of Q numbers, possibly more sophisticated, but there’s no reason to drop that idea.

– I agree with you.

– So I’m uneasy, because again, we know from simple experiments, if you want to destroy one bit, one quantum bit of interference, if you have two slits, two doubles, the famous double slit experiment, and the particle goes through both slits, according to quantum mechanics, at the same time. All you need to do is another qubit, which actually monitors which slit the particle goes through. And that destroys it, there’s no intelligence needed here at all. So one qubit worth of entanglement and shorting it, it’s all you need. And that’s where probability should already start to appear, as you say, in each of these subsystems separately. So, I am this setting. I would like the Born postulate to follow more naturally from quantum mechanics, but I don’t think we have a way to do that.

– And so, from what you say, I guess you’re probably not a great fan of the approaches that so-called, are more epistemically oriented, that it’s all about human knowledge. That’s what the probabilities of quantum are. And when we learn something, well, you update your probabilities because your brain now is aware of things. That seems pretty far removed from the fundamental,

– Far removed, far removed. I think that language is useful. I tell you when it’s useful. Yeah, this is almost the other extreme of what I would like. It’s all in your head. And it seems to me you can defend it philosophically like you can defend any form of idealism. I mean, there’s no way you only– You only ever know things inside your head. Right.

– Indeed. There’s nothing I can do to prove to you that it’s not all in your head.

– Right.

– Right? But the place where I like this kind of discussion of updating is when people talk about entanglement and they say, you know, spooky action at the distance, that was actually one of Einstein’s complaints, where he says, look, I measure this particle, and suddenly the other particle assumes exactly the same state. And I could even change how I measure this. But magically, the other guy, if you have the right state, will assume the the same state. I think it helps to think of your measurement then kind of classically probabilistically, even though there’s just an approximation, in the sense that you could say, well you’ve learned now the state of this system and you knew prior to that that they’re perfectly correlated.

– Relationship, right.

– Therefore I have to update for myself the other system. You yourself incidentally, who holds the other system and is not aware of my measurement, you still don’t know. There’s nothing spooky happening. I would have to pick up the phone and say, “Brian, listen, this is the outcome I got.” In which case you update your knowledge. But then it’s all causal and, you know, behaves nicely. So I think that language is okay there to say, well, you know, quantum superpositions, in terms of updating, don’t really add anything to our classical probabilistic story, but that doesn’t mean quantum mechanics is.

– Right. But I have to ask you on that, so does that betray that you have a certain spooky anxiety? In other words, I’ve come to a place with spooky action that I’ve come closer and closer to a perspective, just accept it. This is how the world actually is. Don’t even try. I don’t want to feel like Neils Bohr, not that I’m of his stature, you know what I’m saying? That’s not what I’m trying to say, but I don’t want to feel like I’m avoiding the difficult questions. But I feel like, especially as we’ve gone on in string theory, we found examples of yet more severe non-locality, as this is called, and we feel something here apparently affecting something over there. Maybe there really is fundamental non-locality in the workings of the world and we don’t even need to rely upon these various crutches or copes to make us more comfortable.

– I love that idea, actually, what you’re suggesting. I like that because I always imagine that if some of these experiments fail, quantum mechanically speaking, that we’ve been discussing, I think it’s gonna lead us into another theory, which actually is more general. I don’t think it’s ever gonna be a return to some kind of classical physics and classical reality with no collapses. But it’s gonna be a theory that’s much more general in the same way that quantum mechanics is a generalization of classical physics, and then quantum mechanics will be a special limiting case of that. So I’m very sympathetic to these views, actually. And in that case, the non-locality you’re talking about, I would say is a genuine non-locality. Which we could probe, you know, at certain small enough distances, for instance, it could really be that you do something here and something over there changes instantaneously, let’s say. And I think the question for me maybe would be how do we test these things? That would be really exciting to probe that.

– And so getting back to tests, you know, in your recent book, I think it was, you may have spoken about a handful of tests, one of which was the one that you described before. What other, like among those, what do you think is the most likely of those tests to be realized in the not, in our lifetime, say? And can you just describe what it would do?

– Yes, I think so. I think there is a handful of, I kind of divided them arbitrarily into five different groups of tests. But they all have many things in common. Of course you’re probing quantum mechanics in the microscopic domain. So I think another culprit for collapse, and I think historically that was maybe there even before gravity, would be living systems. And I think people like Vigner, von Neumann even thought that either living systems or even some higher properties like perception, consciousness–

– Somehow different.

– Somehow different. And they would be responsible for this collapse. So I’m very fond of experiments that are going deeper into chemistry and biology. So for instance, one thing I’d like to be able to do is show that you can quantum entangle two living systems. We’ve done this with inanimate matter, but really going in the direction of Schrodinger’s cat. But of course, cats are far too complicated to do anything with.

– But Schrodinger’s bacteria or whatever.

– Schrodinger’s bacteria is exactly the way to do it. So I think what we want to do is take two tiny optical microcavities. So again, we are talking about mirrors that are a micron apart, and they exist there simply to confine the system and to confine light to be able to control it properly. And then have these two microcavities and a bacterium in each of them, and then put them in different arms of an interferometer. So literally take a single photon, a particle of light, split it to the beam splitter, and then get it to be absorbed by one of these bacteria and the other bacteria. So you’re creating kind of excited ground and ground excited state of these two bacteria, which to me sounds phenomenally interesting. And you know, would that, so a person like me, again, thinks, well it is phenomenally difficult to make that experiment and to control it, but would it really contradict anything that we know about living systems? You know, where is the contradiction now? Even if I prove to you that they’re entangled. So what, you know?

– Right. I mean I think, you know, again, as we discussed our prejudice is quantum mechanics rules. So I think quantum mechanics would persist, but if you could do that, that would pretty sad. I mean, do you know the answer? I don’t know the answer. What is the largest ingredient that has been used in a double slit Like experiment?

– Yes. I think it must be a good friend of mine, Markus Arndt in Vienna. So this is from the group of Anton Zeilinger, the Nobel Prize winner a couple of years ago. And I think it’s exactly these kind of experiments that the group does tests quantum mechanics in all of these extreme situations. I think he took, I think it’s a molecule that has about hundred thousand. So it’s still tiny compared to the scales.

– The cat.

– The cat, definitely. Tiny compared to, even to compared to the, we are still many orders of magnitude away from a virus, even if you’re thinking sending a virus. But I think he’s quite confident that we are going in that direction.

– But you say 100,000.

– 100,000 atomic units, let’s say. So it’s a protein which is as large as that, 100,000. Again, we, you know–

– And obviously it works. We’d have heard If it didn’t work.

– And it works. Indeed. That’s right, we would hear. And I think even people from Vienna who are more like possibly more sympathetic to these kind of semi, the more Copenhagen view of quantum mechanics and.

– Oh, totally. I mean, Zeilinger, we had dinner with him, a few, I don’t know, few, with the COVID, everything, you gotta add like five years or something. But maybe it’s 10, 12 years ago or something. And his view was, of course it’s Copenhagen. Which I found interesting. Because to me, and again for the audience, the Copenhagen approach is more a set of rules to follow. It’s not even a complete theory. Because if you interrogate what happens between the rules, the theory just doesn’t tell you.

– It doesn’t tell you anything.

– Follow these rules. And that’s obviously, we often teach our students that way so that they can solve the problems on the exam. But, you know, the more precocious of those will say, “But wait a second, what about this, that or the other?” And honestly, we don’t really have an answer.

– No, we don’t have an answer.

– And so, but nevertheless, his view was stop talking about these issues. It’s just the Copenhagen approach, and just accept it. Which is a very interesting way, you know, for someone who is really probing the essence of the theory.

– It’s very interesting. And I think I’ve discussed it also with him many times, and I think he gets frustrated. He even asked me–

– That’s right. It feels like he gets frustrated.

– What would it take to convince you that, you know, the world is not like that? And I think the interesting thing is that all of these experiments that they’re doing, they would really be betting themselves on the quantum outcome. And that’s interesting. So somehow if you ask him, is quantum mechanics therefore not a universal description? He would say, “No, no, it’s universal.” And yet somehow I need to sneak in this classical, classical world and classical observer. Even though he doesn’t think that there is a hard boundary. So, maybe that’s a good way to talk about another experiment that I envisaged that I think in the, this one is much harder. And it seems to me this experiment has a chance to change minds of people like him that at least, you know, at that–

– That’s a high bar if you can get there.

– It’s a high bar. I don’t know how difficult, maybe he would just say, “This experiment will never be done.” Although, now with AI, I’m not so sure. So the idea would really be to try to convince you directly that when you’re observing a superposition, even with binary outcomes, only two branches of the universe, that you really exist, whoever is you now.

– In the two.

– And you can see in the two. So you see, even our grammar, our language now becomes insufficient. I think it would be nicer to have a different language, which even describes more directly what this is. So, imagine in the Schrodinger’s cat experiment, if instead of a cat that was a human, of course, and not dead or alive, but do you see bit value, zero bit value, one, you know, is the electron here or there? I think you could, in principle, probe, you could ask this person whether they see a definitive outcome. That’s the key thing. So there could be someone else who is outside of the whole laboratory who prepares this, you know, has, this is–

– Like a Wigner friend.

– Wigner’s friend. It’s exactly what Wigner envisaged. So you have a person inside who is in an entangled state as far as the outside world is concerned. But in each of the branches, they feel quite confident that they’ve seen only one outcome. The electron is on the left, the electron is on the right. And now how would you test whether this really exists at the same time? Well, you could have an external experimentalist who puts a piece of paper inside which says, do you see a definitive outcome?

– But probably don’t tell me what you think you saw.

– Don’t tell me what you think, because then you really become another friend of Wigner’s. You’re right. And so if this person says yes, and again, quantum mechanics would suggest in both branches, the person would say, yes, you know, one is on the left, but I’m not telling you this outcome. So that tells us that the observation is definitive in both branches. And now to really convince you that the state is quantum mechanical overall, you would have to bring them back and interfere them. And that’s the challenging part. So that’s probably where, where people like Zeilinger would say, “Well that’s really unrealistic to do with a.”

– And just so people get a sense of how unrealistic or not, to bring them back together would be to kind of un–

– Interfere them.

– Undo.

– Undo. Exactly right. And that’s what’s difficult.

– And so maybe just describe how difficult you think that is.

– I think this is phenomenally difficult because you have to, when we do this with two atoms, it’s extremely easy. You can entangle two atoms, which you can think of as one atom is observing another atom. And then you can just apply another laser pulse to actually disentangle them, and bring, and people do this routinely. That’s part of quantum computation. What you would now have to understand pretty well is our perception, even. So, you would have to control the way in which this person inside the lab couples to the electron to make that observation. And remember, this enters our eyes, then it triggers the whole neurological system. Then the brain, you know, starts to think about it. Your whole body responds to the observation, and everything gets affected rapidly. So I wonder whether if we had a handle on the very beginning of that process.

– Ah, just enough of it.

– Before the, just enough of it. I don’t think our neuroscience experiments are sophisticated enough to do it, but you know, I wouldn’t rule it out completely that this could be done. But, so you want to kind of do it while it still hasn’t spread and decohered across all of these degrees of freedom that you cannot control. But basically, what you would have to do ultimately is undo the whole observation. Both of these would have to be reunited. And then the question you’re asking is, would you get a single outcome? Like in any interference, or would really stochastically now get.

– This or that.

– This or that.

– Right, right. And so, you know, one of the famous philosophical realizations in the early decades of quantum mechanics, whether you believe it or not, it was certainly a lesson that people were trying to spread and absorb is if you ask a question about something that you can’t really do, then you’re a kind of category area. You’re making a mistake, you’re getting caught up in something that isn’t really a question, because questions really only refer to things that can be carried out. Do you worry that that might be the way to think about a so-called experiment of this sort?

– Yes, I understand that you are, I do encounter that, especially when you speak with philosophers, for instance, they would say, well, there’s probably a disconnect between these two domains. And so, you’re even making a category, as you say, mistake when you are connecting these two. Again, you cannot answer this with any experiments at present, but it’s just our physics intuition, right? That we all, we would be really disappointed if we couldn’t describe nature in one uniform way. And that’s why I’m saying there’s certainly at present no reason to give this up and to really proceed to test whether our perception. Maybe we get surprised. Indeed, maybe there is something more to our perception. Maybe we even need different laws of, you know, Schrodinger famously speculated and asked, “Does life require different laws of physics?”

– Right.

– It may be. Who knows? But it seems to me so far that everything conforms beautifully to quantum mechanics. And I tell you another reason why I’m not worried that if we ever are able to simulate conscious systems with simpler computation, so there’s nothing to suggest that even quantum computers with 1000 quantum bits cannot simulate some kind of rudimentary conscious processes. We don’t know, because we don’t really understand it. But if we could do that with 1000 quantum bits, then the experiment that I have in mind could actually be done.

– Right. But the tricky thing there, of course, again, comes back to the same question, you and I don’t know that either of us is conscious.

– Yes.

– So how would you interrogate this quantum system?

– You’re right, you’re right. How can you trust the answers even?

– Yeah, ’cause it could just be mimicking.

– You’re right, you’re right. You’re right. So you maybe have to phrase the question where the output would not be an answer, but maybe some kind of behavioral thing, right? If you could, for instance, this is probably even less realistic now, but if you could make a bacterium go after food and not go after food simultaneously, so you create a superposition of these alternatives, and then the bacterium really splits and goes to, so rather than requiring on some kind of communication with the bacterium.

– You just observe the behavior.

– You just observe the behavior. So maybe there is something in that.

– Right. Who knows, who knows?

– Who knows.

– Now, on a more fundamental, descriptive level, you, I believe, hold the perspective that I think many, I would say string theorists and others in the quantum gravity world are heading toward, that information is actually the basic currency of reality. Can you just give a feel for what that means from your perspective?

– I love that. I love that picture actually. And it’s another one of those visions that hasn’t been realized. I think there is a lot of mileage that We will get out of this view. So first of all, if you look at the postulates of quantum mechanics, they really have a very strong information theoretic flavor. And I think this has been said by many people. You know, you’ve got John Wheeler saying it from bit. I think Weizsäcker on the German side of things, equally, I think was trying to describe reality in terms of information. So, I think that there is a feeling in all of our communities that maybe these abstract laws of quantum mechanics could actually be made more intuitive by writing them in terms of information. So they already look like information, you know? Schrodinger a called the wave function a catalog of information.

– Sure.

– So there’s already that flavor there, even at the beginning. But I think it would be nice, instead of talking about mathematical structures like Hilbert spaces and all of this, it would be nice to actually find a possible information theoretic structure behind it. And I think that’s what’s been driving me and many other people. Of course, you’re always left with the laws of how this information ought to behave. So you enter this debate, what comes first, therefore? Could the information really be more fundamental?

– And that’s really the question I’m asking.

– That’s the question.

– Do you think?

– I think so. I think that’s very difficult at present because we always, especially in my own field, we think about the fact that there is no disembodied information.

– Exactly.

– So whatever you do.

– It doesn’t have to be instantiated in Stuff.

– Indeed, indeed. So somehow for us, physics comes first, and then obviously classical systems obey classical information. Our quantum systems go beyond that and can mimic classical, but they are exponentially more powerful. And now, what would it take? And I really don’t know how to imagine that, is to go one step further and really detach information from this underlying material, substrate, if you like, and not clear how to do that. I think Weizsäcker was trying, he was trying to come up with some kind of axioms that information should actually obey. And then he came quite close to quantum arguing, that look, they look almost quantum mechanical and things like that. But I don’t think we have that.

– Yeah, yeah.

– At present.

– And so, looking to the future, I mean, do you think in our lifetime, there will be experimental evidence where we can argue really definitively that gravity, as we have long assumed, must play by the rules of quantum physics?

– Yes, I think so. I’m very excited. That’s what excited me about this double interferometer experiment. I really think that within the next five to 10 years, we will have that experiment done. And I think it will show entanglement between these masses. And so we will know that, at least at this linear regime of weak Newtonian limit, if you like, of Einstein’s theory, that it really is quantum mechanical.

– And as you look, and I’m not trying to lead the witness here, but as you look at the proposals, I mean, have you spent time thinking about the various quantum gravitational approaches?

– Yes, yes. And it’s interesting. I like that not only because of the practical side, how you would discriminate them? But they actually tell you different stories.

– Yeah, they do.

– And that’s the beauty, isn’t it? That in one case, you know, canonical approaches would say, well, it’s these different components. You should be thinking about the metrics, you should be thinking about these Christoffel symbols. So it’s still very abstract, and they are your fields and they become Q numbers instead of C numbers. And that’s really the underlying elements of reality. But it would be even more interesting, what we discussed earlier and what you mentioned in terms of geometry, to really realize ultimately that maybe you should go beyond that. And maybe even bits of geometry, which to me sounds almost like science fiction, you know, that you cannot measure, you cannot measure an area and the volume of an object at the same time. That really sounds crazy, but that would be amazing.

– It can’t be versions of that, for sure.

– It could be versions of that. And we don’t know. Maybe it’s not like that. So I, like I said, I always think that this will force us, even if the experiment actually doesn’t work out, my instincts wouldn’t be its collapse. My instincts would be there is a more general theory, actually, and we should now phrase this within this more general opinion.

– Have you looked, by any chance, at any of the work of ‘t Hooft, who’s been trying to imagine that maybe there is a Einsteinian-like classical undergirding to it all and that K quantum mechanics is kind of fooling us into thinking the world is more strange than it actually is?

– Yes, yes. I looked at them briefly. I think I dismissed them maybe a long time ago, only because it seemed to me that somehow you would find it extremely hard to reconcile that with our violations of Bell’s inequalities and things like that. But I think he’s become more sophisticated since, and I think maybe at some small enough level, this could still pass these tests. But my suspicion is that I think any of these returns to classical reality–

– Going the wrong direction,

– It’s going in the wrong direction. They’re problematic, really.

– Right. Now, I tend tend to agree. And look, if we can gain any experimental foothold in the arena of quantum mechanics and gravity, I think that would be an absolutely amazing thing to have happen.

– I agree.

– But certainly, make me feel like the work that we’ve been doing is on a somewhat stronger foundation.

– Yes, you’re right. And that’s what’s exciting to all of us. Because you look at these developments of technologies that are purely motivated by all of these applications, but ultimately we know that we are in this game because we do want to understand reality. And in fact, even with that approach, we know that our technologies ultimately will benefit even from this. Because you’re basically demonstrating a very simple quantum computation using gravity.

– Yeah.

– Which is phenomenal as well.

– Yeah, that would be really amazing. Well, Vlatko, this was a great conversation. Thank you so much for joining us.

– I enjoyed it very much.

– Thank you.

– Thank you.