transcript Valentini discussion

July 21 14.30
Pilot-wave theory: Everett in denial?
Speaker: Antony Valentini
Commentator: Harvey Brown
Floor speakers (in order of appearance):
Bacciagaluppi
Lehner
Janssen
Bub
Saunders
Albert
Hawthorne
Lewis
Loewer
Deutsch

115

Comment by Brown

1. That was a terrific defence of the de Broglie-Bohm picture but I’m still not convinced. I’m going to start by making a couple of concessions. First of all, I think it is fair to say that the claim, the famous David Deutsch aphorism about “chronic denial”, I think that is shaky if you don’t take into consideration, or when you take into consideration, the issue of the non-equilibrium regime. ..the de Broglie-Bohm picture is precisely something that Antony himself has developed, and I think those of us who’ve been defending the Deutsch adage and trying to articulate it have been amiss in not addressing this issue. I’m going to return to this later because I’m not sure actually how shaky the adage is. The other thing that I want to concede is the point that neither de Broglie nor David Bohm developed their theory in response to the measurement problem. This is something that has been well-established by Antony and Guido in relation to de Broglie and I would refer you to the Brown-Wallace paper, there’s a long discussion at the beginning of the paper precisely on this point in relation to David Bohm, partly as a result of information that we got from Jeffrey Bub who of course studied with David Bohm.

2. The Bohm theory was certainly not a result of an attempt to solve the measurement problem, David Bohm didn’t even understand the measurement problem in 1952. I would like to take a quick look at Bohm’s original paper just to fix ideas. I do not want to look at the de Broglie-Bohm picture in it’s own terms. This, after all, is a conference about the Everett picture and I think we should be looking at de Broglie-Bohm from the point of view of Everett. And so I’m going to do this, by way of looking at the 1952 paper, just to fix our ideas. The 1952 paper of course followed the publication of Bohm’s textbook on quantum mechanics which has a very interesting and very extensive chapter on the measurement issue. I won’t call it the measurement problem. This was a very sophisticated account of the measuement process. Although the book defends what Bohm calls the usual interpretation, by which he means something roughly like Copenhagen, he goes absolutely against Bohr’s claim that you can’t have a full quantum-mechanical account of the dynamics of the measurement process. That’s exactly what this chapter is doing. And any reader would be forgiven for thinking that Bohm is perfectly happy with his account of measurement but there just isn’t a measurement problem. How does he do it? I’ll come back to that in a minute. But in his 1952 paper when he introduces a hidden variable theory he has in the second part of the paper, or the second paper in the joint paper, he has a discussion of an impulsive experiment, we might like to think of it jsut in terms, for example, of a Stern-Gerlach experiment, so we have two possible outcomes. If we take into consideration all the measurement apparatus and we think of the detector screen as all part of our configuration space then we end up, says Bohm, with basically two peaks of the wave function, non-overlapping because there’s essentially decoherence, in the configuration space and Bohm introduces what we call the result assumption. The result assumption is the claim that that peak, that hump in the configuration space, the wave function hump in the configuration space, that corresponds to the outcome of the measurement, is the one that’s picked up by the corpuscle…..That’s to say by the configuration that resides in one or the other of these two peaks.

3. Now, what does that mean? The language that he uses, and this is by the way consistent with the language in his book, is that the wave function by itself, each one of those peaks, autonomously has all the structure you need to account for a definite event. After all, if that weren’t the case everything that he wrote in his book a year earlier would have been false. The wave function itself, that peak in configuration space, has all the credentials for representing a definite outcome. Now, in both cases, I think this is a little bit speculative, especially in relation to the book, but I think both in the case of the book a year earlier and in the case of his paper Bohm seems to think there’s only one of them in reality. The reason he thinks this in relation to his book is because actually he uses a rather modern argument, not a very good one, but you see it very often in the literature, to the effect that once the system has interacted with the apparatus, becomes entangled with the apparatus, you look at the sub-systems, you introduce the reduced density matrix, it looks like a mixture, therefore you pick one element out of the mixture because now it’s just a question of providing information to pick out what you were originally uncertain about. So in that case there’s really only one wave packet and somehow you pick it out by just gaining information on where it is as a result of the measurement. In the hidden variable theory it’s the corpuscle’s configuration that picks the special peak in the configuration as the result. But the wave function itself has the credentials, it’s simply which one of them is the right one, which one of them corresponds to the outcome. Now, can this be right? I think, if I understand Tim’s position, this is just not comprehensible. The wave function does not have the structure. It does not have the credentials and, as I said the other day, Tim is not the only person to say this. Peter Holland in his book “The quantum theory of motion” said it very, very clearly. He talked about the particles or the hypothetical corpuscles together with the wave function providing the form and substance of material bodies. But this is very seldom said. It’s almost certainly not Bohm’s own position. Tim may be right, I don’t think he is, but at the very least I think we can say that the issue is not an obvious one. I do not think this is an obvious issue. That’s to say it’s not obvious that that peak in the configuration space fails to have the credentials associated with a definite measurement outcome. And this is exactly the problem with the de Broglie-Bohm picture. If you cannot convince yourself that the wave function is somehow insufficient then the role of the corpuscles in picking out one of them suddenly seems to be, if not redundant then rather mysterious. What is the active role of the corpuscle in this particular case?

4. Now, the reason that we looked in the Brown-Wallace paper at the case of a definite outcome…for example the original spin system’s in a state of spin-up and ………to some point on the screen and then there’s the whole wave-functional description of that situation, was not because that’s a measurement process. It’s because the vast majority of commentators, I think rightly, think that that quantum-mechanical description, which you get by probability one in this case, is all you need to account for the definite outcome. Whether that was the result of a measurement is neither here nor there, it’s what is the ontological status of that final state of the system plus apparatus? If it has the credentials well then unfortunately it’s difficult to see what active role the corpuscles are actually playing. Now, I think it’s important for us to separate ourselves from de Broglie’s original motivations. De Broglie was clearly introducing particles right at the very beginning. And even to separate ourselves from Bohm’s initial motivations …simply to show that some algorithm for picking out one element amongst those in the configuration space could be written down properly and it was a violation of von Neumann’s no-hidden-variable theorem. That was basically Bohm’s position. I don’t know whether he believed his theory initially but that was precisely what he was trying to do, to show that such a theory was possible.

5. Let’s stand back from all these historical considerations. Given what’s been said, what we want to know is: if the wave function has this ontological robustness and it carries the credentials for representing, if it’s peaked, for example, in a suitable way in configuration space, the outcomes of measurements why do we need to introduce corpuscles? Obviously if we don’t we’re into something like the Everett picture. The Everett picture saves the appearances, unless you adopt something like Tim Maudlin’s line. And in that case, if it saves the appearances, one can now appeal to arguments that we’ve been looking at ealier in the conference related to how probabilities emerge, whether or not you take the Deutsch-Wallace representation theorem, whether or not you take something like Wayne and Hilary’s account using a de Finetti-type decision-theoretic approach based on learning from experience in a probabilistic context and so on. Now it’s starting to look as if the Born rule is going to emerge naturally in this thing, in this approach. And that doesn’t fit very well with non-equilibrium statistics. So, was the original claim that de Broglie-Bohm was Everett in denial ….that it was shaky in relation to the equilibrium statistics? Well, it depends on how well you can motivate the corpuscles. I wouldn’t bet on them.

Valentini
6. We seem to agree that if you have non-equilibrium then ….[noise]… You said we should look at it from the point of view of Everett, I still think it’s important to evaluate theories in general on their own terms. Bohm, I agree, there are some shaky statements in his papers. I think he compares the wave function, for instance, to the electromagnetic field. I think he had some wrong ideas that de Broglie didn’t have. Separating everything from history, you were saying at the end, look, let’s think now, forget all the history, what is the best theory? How do we motivate the corpuscles? One can ask these questions. One can and should also ask how do you motivate eigenvalue realism? How do you motivate thinking that I should represent things in terms of this psi because in some cases, if it’s localised, it looks like a pointer, but we see that really in general in realistic systems it’s highly de-localised. The discussion then becomes something very different. I’ve addressed – talke pilot-wave theory seriously, is it many-worlds in denial? Now we’re asking – well, okay, maybe it isn’t but now let’s move on and ask which theory’s better. I don’t claim to know which theory is true, but I do think that Everett has been based on this peculiar assumption, taken from classical language that ….and so on. And no doubt you will have some arguments that pilot-wave theory started with some peculiar assumptions and we should continue this debate, and here we are.

Bacciagaluppi
7. I’m going to propose that the match is even for the following reason. If the pilot-wave theory takes the wave function seriously the argument is these wave function components have the correct structure, the right credentials for counting as independent reality, measurement outcomes and so on. And that is a problem for the pilot-wave theorist. But if the Everettian entertains the possibility that there might be corpuscles that will invalidate the assumption of Equivalence. Then, if you’re comparing wagers, there will be a difference between physical states given one wager or the other one and you won’t be able to apply Equivalence unless you already put in the assumption that observers are going to be indifferent as to what the corpuscles are doing. But I guess that’s begging the question against Bohm.

Valentini
8. Arguments about deriving the Born rule – de Broglie-Bohm in non-equilibrium opens up all new things, non-contextual probabilities and so on. There’s work to be done in thinking about these questions. To me the important point is that these are distinct physical theories and if people want to argue backwards and forwards about which one they prefer for various reasons and assumptions I think there’s a whole range of things that one can say.

Brown
9. I’d just like to make one remark in relation to this. It’s certainly true that if you introduce corpuscles then the logic behind the equivalence principle and the representation theorem fails and this is precisely why you allow the possibility – you can’t rule out non-equilibrium by logic. There may even be other reasons for having doubts about the representation theorem such as that maybe it only makes sense when the agent is on a branch that has the Born rule in the agent’s memories in which case it’s only applicable in precisely the cases where you don’t need it, at least if you adopt the line that Hilary and Wayne are pushing. But the issue that I was trying to raise was why introduce the corpuscles?

Valentini
10. Why cut them out? Why did Schrodinger cut them out?

Brown
11. Because the bare theory saves the appearances.

[new question]

Lehner
12. What I do think is remarkable about Einstein’s reaction to Bohm is that he was not at all impressed by the Bohmian proposal and I think we now know a bit better why that was, why one should think that would appeal to him, that already in the early twenties he tries something similar and what he found is exactly what Harvey was also pointing out, that the more you think about it the more useless this particle part of the particle-wave duality becomes, right. So in the end I think it’s just a good physical point against the Bohmian story that this particle has none of the things that you expect a physical particle to have and he basically had to claim that physics over the last seventy years has measured all the wrong quantities, that all the physical quantities that should interest you are not the wave quantities that we’ve been measuring but the particle quantities and that physics is basically deeply misguided about what the world is like and I don’t see how that is better than the Everettian claim that our concensus is deeply misguided about what the world is like.

Valentini
13. So, what you said about Einstein, historically, is just wrong. Einstein was developing, never published, guiding field theories for photons and he didn’t publish them because the guiding fields were only in three-space, the correlations were not strong enough to give you energy-momentum conservation for single events, which Einstein believed must be right, unlike in the BKS theory which was a sort of theory like this. There are various other things that are – Wigner reports of a symposium in 1925, Einstein’s doing a similar thing now with electrons too with de Broglie guiding waves; again, they were three-space fields, he couldn’t get energy-momentum conservation, that’s why he abandoned it. There’s also unpublished work by Einstein about looking at a mirror in black-body radiation and he’s assuming that photon impacts at distant points are statistically independent and he realises that the mirror is going to develop an unbounded mean velocity through Brownian motion, which he rejects. Generally in the twenties there’s a whole – Don Howard has written on and uncovered a lot of these things – a wonderful paper about the pre-history of EPR. Einstein saw in the early twenties that if you took quanta seriously that they were leading you towards some form of non-locality, he again saw this with Bose statistics. So he drew the conclusion around the mid-twenties that the basic concepts of quantum theory must be wrong, because they lead you to non-locality and he saw this even more strongly in 1927 at the Solvay conference where his objection to quantum theory was essentially the EPR argument. There’s and extensive discussion of this in our book. And the reason, to my mind, and Don Howard also makes this very clear, the reason Einstein thought that the whole quantum structure was just bad is because he saw it lead to non-locality and when Bohm, in 1952, made the following move which appeared to be like this: that I take the quantum formalism and I reinterpret it and add something to get a theory, Einstein thought that, look, the quantum formalism is not a good point of departure. He’d already decided that twenty-five years earlier and I’m quite convinced that that is why the Bohm theory was no good. It was starting from something that was just bad. But now we know that we have to have non-locality, at least if we’re in a single world, so all of this train of thought that Einstein had from the early twenties while he was assuming locality, you take that assumption away then this line of thought that he took we would now say is wrong.

[new question]

Janssen
14. I very much enjoyed especially the historical part of your presentation and in showing where this…..stuff came from. I was a little disappointed with your last slide where you tried to do something similar for Everett and there I’m afraid, pardon me, we haven’t sorted this out, that in the end that history is going to look as bad as the history that you criticised the de Broglie-Bohm….right, so the thing I’d really like to say sort of illustrates how badly………and what more work needs to be done to sort out these things. So in that vein I want to make a few comments on the little strand that I’ve?.been looking at. This has also been mentioned a few times by David Deutsch yesterday about quantisation being – you quantise a classical field as somehow being sort of dubious state?..of relying old classical theory. Now, I think that’s really unfair to the other great pioneer here of quantum mechanics, namely Heisenberg with his kinematic ..[German term]..where the diagnosis of the problem was that there’s nothing wrong with the dynamics of classical physics, there’s only room?.for kinematics, right, and that’s where the….interpretation…of variables….operators comes in and so……is the best example, you don’t throw out masses of equations you just reinterpret them.

15. The example that you gave………string, where you were saying “Well, who in his right mind would say it’s the sum of all these different modes”, well of course that was the very first example of quantum field theory by Pascal Jordan in the ……., right, …exactly he thinks of the ontology as excitations of the field and where the main point that he makes is that, look, by doing that I can get all the behaviour of both waves and particles out and that is a big advance over these ideas that were floating around in the early twenties of which de Broglie is an example, namely, we have this strange dualism where you have fields guiding particles. As you said, this on the back of the BKS?., this is something that Einstein….about. My suspicion is that, certainly for the Copenhagen -…..camp, that a big reason for not liking this de Broglie type of story is that they have come up with a way of having a sort of unified dynamical scheme that could account for all the wave and the particle aspects. And I think….historical…throw up many more interesting lights on this whole story.

Valentini
16. There’s an interesting history about what Heisenberg did, but the particular point that was important for what I said was that Heisenberg and Bohr made an assumption which, let me give an example. Let’s say I want to measure the momentum of a particle. In classical physics, if the particle is in a box I can open the box, I can do a time-of-flight measurement, bang, I find the particle here, distance divided by time times the mass, that’s the momentum the particle had in the box. This kind of thing runs though everything that Bohr and Heisenberg are doing….

Janssen
17. That’s all later, okay. I’m talking about 1925….

Valentini
18. Now, there is no rational reason why, if I have a non-classical particle that to measure its momentum I should do a similar operation. The pilot-wave theory gives you an example, it may be a wrong theory, of showing you that that is in fact not the true measurement.

[new question]

Bub
19. …by the way it was a terrific talk. Bohm always used to say that no one is interested in doing experiments just to find out the position of a particle……[indistinct aside]……………….what we’re really interested in is precisely …………spectra that is for example……….that is, measuring the eigenvalues of…..operators, and the story in terms of the position of a particle is very far removed from that. It’s different in classical mechanics because the observables that you’re looking for are just functions of position, assuming the whole position story and you’re really talking about values of ……

Valentini
20. Well, I agee, and look at what actually happened. The data that people were looking at were spectral lines, atomic energy levels and very much – that was early..in the twenties most of the experimental data that people were looking at were scattering experiments. You know, all kinds of …the cross-sectional area varies with angle. The ……effect stunned people: that a very low energy electron would pass almost freely through a gas, and how do you explain that, and it turned out that Alzasser?..if you have a long de Broglie wavelength you could explain that the scattering cross-section went almost to zero….certainly they weren’t measuring the positions of particles…But it was about electrons, atoms, then it became about photons and – I don’t see how that …[indistinct ruminating ]…clearly the data you see, they’re not just directly in terms of particle positions, there’s a complicated story about the experiments.

[new question]

Saunders
21. I share your view that the non-equilibrium, the charge really takes on a different character. But sticking to the equilibrium case, I’d like a clear verdict from you. Sticking to the equilibrium case, as many Bohmians do, it seems to me that, to come back to he rhetorical issue of does one look at Bohm through Everettian spectacles or not, if it were the case that an Everettian is making some special postulate, as it were, alien to the formalism, alien to standard scientific methodology, then of course there would be something wrong with the accusation, it shouldn’t anyway disturb the Bohmian.. But if it is the case that the Everettian is not using any special postulates or assumptions, if the methodology of interpretation is standard scientific methodology, and therefore that that standard methodology applies to the Bohmian equilibrium case, then, if all that goes through, it seems to me that the argument is really compelling that the trajectories are epiphenomena. If it doesn’t go through then of course that’s a different issue. My own strong sense is that either you take Tim’s line, that it’s just conceptually incoherent to talk about wave functions without the configurations as additional ontology, or you say that the Everettian methodology doesn’t go through, or doesn’t deliver, or that it makes use of tacit special interpretative assumptions. If you can’t say one of those things then the equilibrium Bohmian I think really is in trouble - and I want to know your view on that.

Valentini
22. ……...[indistinct preamble]…difficult to come to a watertight conclusion about that but I have an earlier problem which is that to me the equilibrium de Broglie-Bohm, and it’s a bit like what Itamar was saying on Thursday, if you only ever have this thermal equilibrium, you never have access to some more detail about molecular motions or more detail about the de Broglie-Bohm trajectories, I mean, look, here we have a theory where there are two fundamental equations of motion…….Two equations of motion, the Schrodinger equation and this equation of motion for the trajectory. Now, the details of the trajectories are almost completely washed out in equilibrium. All you see is that the velocity law must preserve the Born distribution with time, a kind of very crude average. Now, if that was the world, if someone says to me this is my theory of the world, these two equations of motion plus this distribution which wipes out almost all the details of one of them then to me, scientifically, I’m very unhappy with that. To me, if you take the de Broglie-Bohm theory seriously it is crying out to you that, look, quantum theory is just for a special state, there must be non-equilibrium somewhere or at some time, it doesn’t seem at all like a reasonable way to make the world for a scientist. It’s logically possible but I wouldn’t accept it so I would say that if we – if in a thousand years we’ve measured everything in the early universe, everything coming out of black holes, everything that one might conceivably think’s a place where there might be non-equilibrium, there’s no non-equilibria, then I will rethink.

Albert
23. …relevant to this, sort of intended I guess as an answer to Simon. Look, I don’t think the two options you laid out are the only two options. I think one of the things that was enormously helpful about John Hawthorne’s talk today was precisely the description of a very reasonable middle ground. If we have very good scientific reasons for believing that such and such is the basic ontology of the world, okay, then we’re going to look around in the world and say, well, I guess those must be the tables and those must be the chairs because that’s all we’ve got. But if we have reasons for entertaining a different ontology all of a sudden, and ontology with extra things there that aren’t in the first ontology I was considering of course it may be eminently reasonable to say, oh, no, those are the tables and chairs, we don’t have to think of these as the tables and chairs any more and it just doesn’t follow at all, and that’s what seems to be the basis of all these arguments, that once you’ve granted, subject to the constraint that this is the basic ontology, well I guess those are going to be the tables and chairs in this ontology, that that somehow commits you to sticking with that in alternate circumstances where you’re considering different ontologies where there are much more plausible, much more credible candidates for what the tables and chairs are. So, you’re pointing to a dichotomy: either you have to reject that in a pure wave function monist theory and a collapse theory, that is, in the simple case, where it’s an eigenstate, either you have to reject that that’s a table or a chair or your going to have to grant that inventing the corpuscles does no good. That seems to me an absolutely false dichotomy.

Saunders
24. Look at Peter Holland, look at what he says for the deterministic case, that it is unintelligible that it can act as it does….

Albert
25. When you say “act as it does”?

Saunders
26. That it can be a basis for….

Albert
27. Oh; it’s not unintelligible to me. But that doesn’t mean I don’t see the appeal of – you know, it’s one of the things to weigh in the balance. It’s not unintelligible to me at all but it would be mad to go to, say, denying what Tim says along the lines of, look, it’s just so much more transparent in the case where you do have the corpuscles. I don’t think I would go so far as Tim in saying that it’s unintelligible without the corpuscles but there’s no question that that’s a point, and there are points on both sides, that’s why it’s a hard problem. But to think that granting that commits you to denying that there might be cases in which the credentials are something else, like John says, swamp the credentials of the wave function to be the tables and chairs, I don’t see how that follows at all.

Hawthorne
28. I want to underscore this by making it vivid because you’ve really been on a total looser with this credentials argument. If you try to bring it against, say, the ontology that Tim gave, where remember that ontology you’ve got a configuration space and a four-dimensional space and certain fundamental relationships between them. Suppose that is the fundamental structure of the world. Forget the issue “is it justified?”. But one thing you guys are trying to do is say, even ….ing ourselves that, it’s still: the credentials have been earned for grooves in configuration space deserving the label four-dimensional spacetime. If, in reality, there is this thing that’s just straightforwardly, naturally associated with the language of four-dimensional spacetime, on no plausible metasemantics is any groove, even a halo groove, in configuration space deserving of the name of four-dimensional spacetime by comparison with this 4D thing that Tim’s …….There’s just no way you’re going to win that argument.

Brown
29. I’m going to appeal to what I consider to be the best principle of metaphysics to pit against your metasemantics. Ockam’s razor. You have a formalism which is shared by two theories, so there’s a dynamics associated with some fundamental element of reality, but one of the theories adds something else. Now the question arises: can you save the appearances in both theories? There is no concensus on this but most commentators would say yes.

Albert
30. Not most of us at this conference

Brown
31. So it’s not an obvious issue. Now, there is no law as know, in the history of physics, that says that spacetime or indeed even space, has to exist in the theory at some primordial level. Why does it? We would be perfectly happy in physics if everything we see around us, in a particular the three-dimensional and maybe even four-dimensional aspects of reality, was an emergent property out of something that is fundamental. So, given that saving the appearances, well-defined dynamics, all these things are a property of the first theory, why in your right mind would you introduce extra structure?

Hawthorne
30. There are two questions. One question is: is Tim’s ontology to profligate. Maybe yes. But there’s another claim that you have definitely been making in the literature which is the conditional. Even his ontology’s true, grooves in configuration space are deserving of the name four-dimensional spacetime. I am challenging that.

Albert
31. Things are being presented as if they were exactly two options whereas there are in fact a continuum of them it seems to me. It doesn’t follow from my saying – if I’m agreeing to play the game, okay, there’s only the wave function. Tim is just going to refuse to play that game; that’s not my position. My position would be, okay, I can see how to make sense of that. I guess, if there are good reasons for believing that, these must be the tables and chairs. But if I’m now entertaining a different ontology with, say, corpuscles in it, of course I’m going to withdraw that claim because there are much better-credentialed candidates around for being the tables and chairs. So I’m not going to say there’re tons of tables and chairs here, I’m going to say the tables and chairs are made up by the corpuscles.

[new question]

Lewis
32. I just wanted to ask about the difference between mathematical and ontological alternatives that – there is prima facie a difference between your motivating examples like the electromagnetic……field and the test particle and the pilot-wave case. In the EM?..case there’s the test particle and there’s nothing evolving along the other possible trajectories that the test particle…..along. Whereas, modulo Tim’s worries about the relationship between configuration space and three-space, in the pilot-wave case there is something evolving along those.

Valentini
33. No, I’m sorry, I disagree. In the case of the test particle in an electromagnetic field at all points in space there is something that is evolving, there is a field everywhere and there’s this particle moving. Now, you could in the structure of the field identify a mathematical velocity field, a trajectory that’s part of……… And it’s the same in pilot-wave, there is a field over space, there is one particle moving and it’s just – you can look at the gradient, the phase of the field elsewhere and imagine that if the particle had been there it would have had this velocity, just as it would have had from the Lorenz force law, it would have had this acceleration in the electromagnetic field….I see it as essentially the same. I don’t know if I’m missing something here.

[new question]

Loewer
34. In order to argue that Bohm is Everett in denial….thinking that whatever it is that the mathematical apparatus describing ..Bohm is the same sort of thing as you think of as describing many-worlds but at least some of the people who are Bohmians think of it really differently from that. They think of it as describing something that’s like a law. That is, determining how particles..[interference]. Now, I don’t know exactly what the ……..a law is but I guess whatever the structure of a law is it’s not the stuff that material objects are made up out of and so this looks like really and alternative theory. And not one in which just has a ghost….ghost particles.

Valentini
35. I would agree with the first bit of what you said but the second bit about – there are some people who claim that wave function is like a law, I think that’s just bad. It’s as bad as saying that the electromagnetic field is like a law and that, you know, the ontology’s just charged particles. I think that view is bad.

Deutsch
36. The reason it’s bad, I would say it’s bad because this law contains people walking around and interacting with each other and affecting each other and I’m surprised that no one has yet mentioned quantum computers in this. I don’t think that measurement theory today is based on this classical assumption that if P1, Q2. It’s more based on the quantum theory of computation. You imagine a computer that can be programmed to simulate a tiger that’s going to do something like eat food, with classical computation but it’s a quantum computer. And then you set out in a superposition of different initial conditions with the food in different places, then if you say that the first computation contained (and you’re committed to the wave function being ontological) if you say that the first case contains the simulation of a tiger eating food in position A then I don’t see how you can avoid saying that the second place contains lots of simulations of tigers eating in all different places simultaneously.

Valentini
37. As I said, from the pilot-wave perspective you can point to mathematical structure in the pilot wave that you could identify with alternative evolutions. But you’re not driven to give them an ontological status

Deutsch
38. ..are you saying it does not contain those evolutions?

Valentini
39. No, it contains a complex-valued field on configurations space which contains..

Deutsch
40. No, in reality..

Valentini
41. Which is a reality. And if you look at the mathematical structure of that wave, yes, it does contain mathematical simulations

Deutsch
42. So, in reality it contains running programs simulating tigers.

Valentini
43. In reality it contains a complex field evolving on configuration space which you can decompose in certain ways and regard as, well, here’s one piece of this..

Deutsch
44 . There’s your Everett in denial.

Valentini
45. No.

Deutsch
46. It exists on the one hand….

Valentini
47. No, it exists as a complex-valued field. I mean, like the string exists as a displacement; it does not exist as a superposition of this plus that plus that; that’s mathematical structure. And this is an important point because in your book you made a challenge: well, how do you explain quantum computing in a theory which doesn’t have parallel worlds? Well, look, first of all, even within the standard quantum theory community people have questioned the degree to which it’s superposition. Some people like Richard Josa..say really it’s entanglement. It is controversial.

48. But second of all, here, I agree that there’s an ontological complex-valued field in configuration space and if you look at this theory, it’s a non-local theory, it’s a contextual theory and there’s no doubt that it reproduces the empirical predictions of quantum theory for experiments, for quantum computers, and I’m sure that if you look at specific algorithms and you take into account that there is an ontological pilot wave in a higher-dimensional space, that it’s non-local, there’s entanglement, there’s contextuality; there’s plenty of room for understanding how quantum computations work just as there’s plenty of room for understanding how many other quantum experiments work. If there were no ontological pilot-wave in configuration space then one might worry, but…