Audio Player
Starting at:
Chiara Marletto: Constructor Theory, Ghost Particles, and New Form of Science
January 9, 2024
•
2:08:41
•
undefined
Audio:
Download MP3
ℹ️ Timestamps visible: Timestamps may be inaccurate if the MP3 has dynamically injected ads. Hide timestamps.
Transcript
Enhanced with Timestamps
292 sentences
18,755 words
Method: api-polled
Transcription time: 126m 27s
What is Constructor Theory and how is it different now than in its original formulation by David Deutsch? So Constructor Theory is a new way of formulating the laws of physics and this was originally proposed by David I think back in about 2011, something like that.
as a new mode of explanation. So he wrote a paper that had a very strong philosophical component which laid the foundations of the theory in the form of a program in a sense. And the key idea there was to modify the way we formulate laws of physics and so the fundamental laws of physics. So instead of using things like
Dynamical equations, so laws of motion and initial conditions, which is what most fundamental theories do. Switch to a different mode where the basic fundamental statements are constraints about which transformations can be performed and which transformations cannot be performed and why. So what is possible and what is impossible?
And then consider dynamics and initial conditions as kind of emergent consequences of these principles. So it's really like a switch of perspective into thinking what is what is the fundamental element in a physical theory.
And this was there. I think that's the key idea. And David was really inspired to do this by the quantum theory of computation, which is a theory that he himself pioneered in the eighties when he with other people proposed the idea of a universal quantum computer there. I think it's really important to in that theory.
To think of what can be performed by a universal Turing machine and what cannot be performed by it under given laws of physics. So in the case of classical physics, you have a classical universal Turing machine that can do certain things and not others. And then you have a quantum universal Turing machine, which uses the laws of quantum theory, not classical physics.
And there you have new different modes of computation available. And this was a key insight in developing this idea of the universal quantum computer and construct a theory. One way to see it, and this was already there in David's paper, is to think of it as a theory of a more general programmable machine that is even more general than a universal computer. And this is a universal constructor.
So a constructor is an entity that can be programmed to perform a number of tasks that are not necessarily computations. So you can think of a heat engine as a constructor, if you like. A 3D printer is a constructor. Anything that can be programmed to perform a given physical transformation, you can think of it as a programmable constructor.
It just has to have this property of being able to do the transformation once and then keep its ability to do it again. So that's the key feature of the constructor, which makes it different from a different system that can just simply perform the transformation once and then maybe be destroyed or whatever. And a universal constructor is the most general programmable machine that we can think of. And this is what the physicist John von Neumann
Thought of when he was imagining the ultimate, you know, the most general program machine that could be built by by by humans in a sense. And constructed theory can be thought of as a way to generalize the quantum theory of computation to cover these machines that are more general than computers.
And this is somehow a completion of what phenomenon had in mind, because phenomenon had this idea of the universal constructor, but then never really deliver the physical theory of these machines. Whereas we are hoping with constructed theory that we will be able to deliver a theory of these machines. At the same time, also deepening our understanding of physical theories, because
When you understand what are the fundamental limits of the universal constructor, so what is it that it can or cannot perform, you've also expressed what are the possible and impossible tasks according to the most fundamental laws of physics. So in a sense, studying the universal constructor and studying what is possible and impossible under the laws of physics is the same. And this is a key insight in David's paper.
There are lots of other things in that paper, I think, different ways of thinking about constructive theory as a way to expand on complexity theory and chemistry and thermodynamics and so on. But back then and I, well, a few years after, I think later, something like 2012 or something, I was doing my PhD. Back then, we didn't have
Any specific application of the theory so it was more like a program right and what happened between then and now and let's say what I was really interested in when I started working with David and then kind of develop various things on my own was that I I like this idea this new switch this this sort of switch of perspective and I thought it was very promising and then I wanted to find some specific problems that this approach could be applied to
So then I think in partly my thesis and then later on in my research work, I did a few things where I applied this theory to various problems. So initially with David, we applied it to information theory and we found a very interesting way of expressing with this language of constructed theory the
laws, the principles that underlie physical theories of information. So it's this theory that we developed together, David and I was to express what are the regularities in nature that are needed for information to exist and also for quantum information to exist. So these are ways of handling both quantum and classical information in the same theoretical framework. And this is very important for direction of research that I'm really keen on
Nowadays, which is the direction where you're thinking of systems that are in interaction with objects that obey quantum theory, but may not themselves be quantum. Maybe they behave according to a new theory. Maybe they behave according to a post quantum theory. Sure. For example, gravity is one of these objects because
We don't know. We have various proposals for quantum gravity, but we don't know which particular quantum theory of gravity is the correct one yet. And so in that case, it's very important to have a framework, a theoretical framework to handle the situation where gravity that may or may not be quantum interacts with the quantum object.
and a framework that can handle both quantum and classical things, let's say, in the same unified scenario. And that's what the Constructor Theory information can do for you, among other things. So it's one direction. Another direction where we made progress was thermodynamics. So there was an application of Constructor Theory to thermodynamics.
and to expanding on the current formulation of the second law, something that we can discuss later. Right. And then a third direction, broad direction was the application of constructed theory to the physics of life. So there are these issues about how what is, you know, what is the simplest entity that can occur in the universe, which can be considered as alive.
What are the, let's say, essential features of this entity? So does it have to be programmable in some way? Is it a kind of programmable constructor? What's the minimal structure of this entity? And in that direction, I think I applied Constructor Theory to tell us under what are, let's say, the necessary and sufficient conditions for
an entity to be capable of self reproducing very accurately. So just like living things do. So in a sense, you know, when we think of self reproducing entities, we think of laws of biology, but ultimately what we can do in biology is really set by the laws of physics that we have available. So it's interesting from the physics point of view, and especially from the construct aesthetic point of view to ask,
Considering the laws of physics as we know them, what are the minimal features that are both necessary and sufficient for a living system to be capable of self-reproducing accurately? And this is the kind of stuff that Constructor Theory can deliver on. I think I developed this branch of Constructor Theory.
We the view of applying this to the study of, you know, for example, the origin of life and possibly the study of life elsewhere in the universe. So these are, let's say, the three macro directions in which progress on and then David has also worked independently on on there is other things to do with the universal constructor itself. And finally, the there are a few things in the pipeline with
Some collaborators of mine, Maria Violares, who is a PhD student, a default student here in Oxford, who's developed some interesting results about irreversibility. So again, about thermodynamics and then some work that David and I are doing on the constructed theory of time. So this is kind of forthcoming. And then some extra work on the applications of constructed theory to this area where
We have a quantum system interacting with something that may or may not be quantum. And this is something I'm doing with Giuseppe Di Pietro, who is another PhD student here in Oxford. So there's a lot of there's been a lot of work. And finally, there's also been an interesting application of constructed theory to the problem of testing quantum effects in gravity, which is something I've developed with Vlad Kovetrov, who's a physicist here in the physics department. So I think that's an overview of what's going on.
Wonderful overview. Thank you so much. Thank you. The audience should know that there's a book that you have called The Science of Can and Can't, which goes over these topics in an extremely introductory manner to people who don't know even what a Turing machine is. And so I have read your recent papers and that book as well. So I'd like to get into some of the technicalities soon. But I would like you to explain the difference between a Turing machine, a universal Turing machine,
And then a quantum Turing machine and a constructor and a universal constructor. So please. Yeah. So this is a great question because it goes, let's say at the heart of the matter. So a Turing machine is a
programmable um machine that so it's a it's an entity which has um you know you can program to do a number of tasks or transformations and these transformations are computations so uh they are a particular kind of transformations that involve um if you like information variables and the classical Turing machine is a Turing machine that operates according to the laws of physics that
If you like newton discovered albeit a discretized version of those laws but let's say you know it runs on that kind of physics therefore does not have all the new and very interesting effects that quantum theory led us to discover about a century ago when it was proposed
A quantum Turing machine is a programmable computer is a program machine can perform computations that obeys the laws of quantum theory. So instead of having the laws of Newton or discretized version of those, we have quantum theory, which is by the way, the one of the best available explanations of the universe. And I think David told me and I like those, which was quite fun at some point where
you know, telling me about how he thought about this universal quantum computer idea initially was then was that he was discussing with someone. And I think it was it occurred to him that somehow the laws of well, the laws of a standard Turing machine, a classical Turing machine were running on the wrong physics in a sense, because they were using an obsolete type of physics, right? Newton's laws, if you like.
Where is the computer scientist should have looked into something that run on the actual.
I love the physics that are updated now and they obey the sort of scheme of quantum theory. Somehow the idea was to update, upgrade the idea of a classical Turing machine with the right laws of physics, with the laws of physics that we know superseded Newton's laws at the beginning of the last century. Now a universal
Turing machine is a computer, so programmable machine that can perform computations that's capable of performing all physically allowed computations. So not just one computational tool or whatever, but if you consider the set of all physically possible computations under a given law of physics, the universal Turing machine in that specific physics
is capable of performing all of them. So for example, you can think of a computation being, I don't know, an addition, you know, you can think of addition and multiplication. These are two possible computations. You can imagine two specific entities, one that can add things and another computer that can only multiply.
Now a universal, like a more universal machine than either of those is one that is a computer that can be programmed to perform either multiplication or addition. And now if you consider all the possible computations, a universal computer is one that can be programmed to perform all of these computations put together. So it's like a multifunctional entity.
And so provided that you give it the right program, it will perform the right computation. Sorry to interrupt with the computer that someone is listening to this on their iPhone or their desktop. Is that a universal computer? Yes. So that's a universal. It's an approximation of a universal classical Turing machine. Yes. So I think that's what Turing gave us with his ideas was basically the
Model the theory, let's say that that powered all of the information technology that we currently use. And the idea is that the quantum universal Turing machine will upgrade these machines when when they you know when when the universal quantum computer comes about in ways that they can perform new algorithms that are based on quantum laws rather than the classical laws of physics.
And our constructors are simply, so if you are happy with this idea of programming something to perform a computation, which is what Turing machines are about, constructors bring this concept a level up in the sense that instead of just having computations as the transformations that you're considering in the repertoire of your machine, you have any physical transformation that is conceivable. So,
in the case of a constructor is a programmable machine that can perform a given task. The task can be a computation. So computers, Turing machines are special cases of constructors, but constructors can be more general. And typically examples of constructors are things like catalysts, computers, as I said, heat engines, 3D printers,
Also some machines that can be programmed to perform a transformation, a physical transformation, and also have the ability that they can perform it and stay unchanged in the capacity of performing the transformation again. This is very important. This is true for computers too. There are special cases for constructors because, you know, they perform a computation and then you want them to be able to do it over and over again. And I think the ideal Turing machine should be
able to do this indefinitely. Likewise, constructors have this feature of being able to perform a task and then repeat this over and over again if given the right input. And so again, programmable constructors are those that can be programmed to perform given tasks. And then a universal constructor is a constructor that has all of the possible tasks in its repertoire. So you can program it to perform any task that is physically allowed.
And there can be a quantum programmable constructor and the quantum universal constructor and possibly a universal constructor that runs on better laws of physics than those we currently know. So maybe post quantum constructors. And so the idea is always the same. Different laws of physics give you different sets of tasks that can be performed just like different laws of physics give you different computations being performable by a Turing machine.
And so, von Neumann's idea was really to extend the scheme of Turing's to other tasks that are not just computations. So thermodynamics transformations are an example, chemical reactions are another example. And von Neumann specifically was concerned about emulating life. And so he noticed that the reason why he thought of this constructor idea was that he noticed that in the Turing machines model,
There was a gap in the sense that you are not. It's impossible to program a Turing machine, a universal Turing machine, whether it's quantum classical or whatever, to self replicate. So you can program a computer to simulate a self reproducing cell. But if you wanted to program your own computer to create a replica of itself, which would be very convenient,
Okay, great. So how does one make a difference between what's extremely unlikely and what's impossible?
So it's my understanding from what you've just said and from the papers I've read is that the traditional way of doing physics is that you have some initial data set and you evolve it forward. That's like the dynamics. And the constructor way of looking at it is, okay, well, actually, let me go back. What you can do is then look at these laws as some causes that produce some effect. And one of those effects may be that entropy tends to increase.
Okay, so in other words, you can derive thermodynamics from statistical mechanics. So it seems like constructor theory is working backward by looking at the effects and then stating those as laws rather than deriving them.
So when I hear you say, look, entropy doesn't, or the heat engine entropy doesn't increase, but well, entropy is not likely to increase. So at what point do we make a cutoff between the likeliness and saying that something's impossible? This Marshawn beast mode Lynch prize pick is making sports season even more fun on projects, whether you're a football fan, a basketball fan, it always feel good to be right.
Right now, new users get $50 instantly in lineups when you play your first $5. The app is simple to use. Pick two or more players. Pick more or less on their stat projections. Anything from touchdown to threes. And if you write, you can win big. Mix and match players from any sport on PrizePix, America's number one daily fantasy sports app. PrizePix is available in 40 plus states, including California, Texas,
This is a great question.
The so let's first consider the fact that when you have to, when you say that something is possible or impossible, you're not directly referring to likelihood or probabilities for it to happen. So somehow probabilities are not there in the foundations of constructed theory. And they come as say derivative statements, approximate things, but somehow they're not necessarily built. In fact, they're not built at all in the foundations of the theory. And this is a plus in a sense. So when you say that something is
a task is impossible. What you mean is that there is a law that forbids the fact that this transformation, this physical transformation that's referred to by the task is brought about to arbitrarily high accuracy by a constructor. So by this device that
can operate in a cycle by returning the substrate that you gave to the constructor in the right input state in the correct output state. And so if the laws of physics say there is some fundamental limit beyond which we can go as far as this transformation is concerned, meaning there cannot be a full cycle that can operate this transformation and the return itself to the original
state of affairs, just like a catalyst would do. That's the case when the task is impossible. A very simple example is the task of changing the energy of a substrate. So if you want to change the energy of a battery, for example, so you know, from low to high, you need so you can't do it with you can't do it with a constructor
Without any other side effects, because the constructor will have to give up some energy because of conservation of energy would give up some energy to the battery. Yes, yes. Thereby not being able to return itself to the original state of being able to perform the task again to the same degree of accuracy.
So that's an example of a transformation that's actually possible in the sense it can happen. It can occur on the laws of physics because obviously we can use another source of energy to replenish the battery once it's on low. However, this other source will have to be depleted itself, so it cannot be a constructor. So now if you're looking for a constructor that can
Give you energy without depleting itself, you will have to go to a physics where conservation of energy isn't true. And so it's not the kind of physics that we believe kind of describes our universe. Likewise, you can think of another example of an impossible transformation. That's the case of in quantum theory. We know that we cannot reliably copy
any two states of a physical system. This is related to the Isenberg's uncertainty principle. The fact that you cannot measure reliably any two observables of a physical system typically is used, you know, we say the velocity and the position of an electron cannot be both measured simultaneously with the same device. Indirectly, this is constructed statement because it says we cannot have a constructor that can copy or
or measure accurately these two variables without changing itself in some way. And so it's no longer a constructor, right? So these are examples of impossible tasks and not that I haven't talked about probabilities there yet.
And now if you have lots of physics that, let's say, tell you a number of things that are impossible, but the rest, they don't constrain then whatever is left is a possible task. So if you, you know, if you think of a way of expressing a theory in constructors, the ethic terms, you will have a number of statements about tasks being impossible. And those that aren't impossible are possible in the sense that then somehow it's, it's, it's allowed to, to bring about a constructor that can
perform these tasks. And there is where you can think of somehow approximate constructors. So when you're thinking of a possible task being realized, performed, let's imagine a possible task is, for instance, to copy the two values of a bit. So, you know, zero and one can be copied. We know it is possible. We do it all the time, approximately in computers.
Um, so, you know, you have zero and one and et cetera. Now, um, the fact that the task is possible is, um, a statement about an idealized scenario where you're thinking, okay, that means that, um, there's a number, there are a number of ways of, um, approximating arbitrarily. Well, this behavior of an ideal constructor can copy zero and one when given them. Yes.
Now, of course, if we look at each particular realization of a copier in any of the computers that we have, for example, they will be approximate. So they won't work perfectly in the sense that at some point they may break down, they may incur in errors, et cetera. But this is simply a feature of the fact that our, you know, we're using limited resources to implement each particular realization of a copier.
However, the laws of physics, as far as we know, don't put any limit on how well we can copy. So it means that for each of these imperfect constructors that are approximate, we can work a bit harder, you know, put a bit more resources into the into the particular device that we have, make it better so we can meet a better accuracy target if we want to.
And so the fact that the task is possible simply means there isn't a limitation beyond which we cannot go as far as accuracy is concerned for this task, as far as we know. So these are very different statements from the statement that something is unlikely. So a transformation can be unlikely or more likely depending on the kind of statement you're looking at. But that would mean simply, for example, that in the standard way of thinking about physics,
there is
If we run the system and those are the initial conditions, if we run repeated experiments, we will see most of the times that trajectory to occurring. But it doesn't mean that necessarily there is a constructor. So this device that can work in a cycle to bring this transformation about. So the fact that the trajectory is very likely doesn't necessarily mean the task associated with it is possible.
Likewise, the fact that the transformation is unlikely doesn't necessarily mean that the task is concerned. So for example, you could say given the initial conditions we have, some transformations that are currently occurring really very reliably in some laboratories, I don't know, in CERN or something, they're very unlikely compared to the standard
physical processes occur naturally in the universe. However, they are possible tasks because we can simply harness enough energy and enough sort of devices that compose the CERN labs, if you like, and we can actually reliably bring those transformations about. So even though some trajectories are unlikely, they can correspond to possible tasks.
In fact, most of the things we do in laboratories, even in quantum computing laboratories, are very unlikely trajectories for certain entities, certain charges or whatever field, etc. And yet we can bring them about really accurately simply because we are following some sort of program to implement these things in the laboratory.
So I guess this is a difference between something being likely and something being possible and likewise unlikely and impossible. I see. OK, let me see if I have the terminology correct. There's something called resource theories and I understand that constructors theory is not a resource theory, but resource theories also deal with tasks and then ingredients that you put together to make some input transformation, some output. Let's say we have this cup and we have a bucket of water, a full bucket of water, and I want to fill this cup so then we can combine
them to make the task full cup of water after the bucket has poured in some water into it yeah but it's my understanding that this bucket wouldn't be a constructor because this bucket runs out of water it's not something you can keep repeating yes yeah okay so then is this realistic then for a constructor to exist there's no infinite bucket of water that exists for instance is that realistic how do you think about this
Well, that simply means the transformations of changing the content of a battery or whatever. If you're talking about a concert quantity, I think in your example, if you like, you can substitute your example with the idea of energy and the battery, because then the conservation of energy is the thing that puts a limit on what you can do. So in the case of the energy conservation and changing the
energy value of a battery. There is a different task that is possible, which is that it's the is the task of transferring some energy from one subsystem to another. So that is a possible task. Meaning there can be a constructed, you know, I think that reliably, you know, if you have a if you have an iPhone or whatever smartphone that's run out of juice,
You can plug it in as power supply. And if you consider the joint system on the power supply and the smartphone, on that system, there is a possible task that can be performed and simply means you're transferring some energy from one side to the other. So it's like, if you want to think about it, not in terms of batteries and charges, you can think of it in terms of a seesaw.
So you have like a seesaw with two weights and they can move like this. And the task of changing the relative positions of these two entities is possible. However, changing the position of one side is not a possible task by itself because you would have to use energy to do that. So you can still talk about the fact that a charger is possible or a seesaw is possible.
By considering the joint system of the thing that you want to recharge and charge the battery supply or if you like in the case of the seesaw of both sides of the seesaw. So it's completely I would say it's completely fine to talk about it in those terms. And in a way it's more insightful because it tells you it tells you somehow
Where is it that the constraints are right so the reason why we need a power supply is because the you know the fact that we care about the fact that battery runs out of juice is simply because there isn't a. What we need to supply the energy from somewhere else once it's gone.
And the reason why we need to do that is the conservation of energy. After all, the fact that overall in the universe, the energy is conserved. And so whenever you change energy in one subsystem, you have to change it somewhere else as well.
And the interesting thing is that you can explain then some limitations of what you can do, for example, with a heat engine or like, you know, any kind of engine that runs on the laws of thermodynamics as we know them in terms of the fact that energy is conserved. So that's the explanatory power of the law of conservation of energy, which you can express in constructive terms by saying the energy, changing the energy of the substrate is impossible.
Um, and so that's the content of the theory. And in a sense, I think unlike what you mentioned resource theory, the, the, the difference between say was constructed theory doesn't and resource theory is simply that, um, well, there's some technical difference, but I think the most important difference is that resource theory is more like a framework where you can express existing, uh, dynamical laws or cement with some symmetries.
in this language of transformations being allowed or disallowed. And then sometimes they also care about the fact that the transformations are performed reliably. And then they talk about the catalyst, which has some overlap with constructors. But the main difference is that they don't have principles of their own. So resource theory is not a physical theory. It's a, it's a, it's a framework to express physical theories. Whereas constructors theory has principle principles of its own. So it's, it's, it's, it's an attempt to have
Actual laws of physics, in addition to those we have currently that can supplement the dynamical laws and tell us more about the universe. So the laws of information are an example. The new laws of thermodynamics that we formulated are another example. So somehow it goes beyond the physics we know currently in the hope of having new predictions, ultimately explanations as well, but also ultimately having informing some new tests that can be somehow testing laboratory sort of changed.
You know, it can provide some sort of extra predictions, extra tests compared to what we currently can do with the laws that we know. When you're repeating a task, are you doing so using a constructor or the constructor? So in other words, are you reusing the same constructor or are you pulling from a different constructor every single time? No, so it's the in the in the ideal case is the same. So so let's talk about the ideal case.
It's a bit like a heat engine is the same constructor. So the idea is you got the same constructor like a fridge and you want to call the. I don't know. Kind of coffee of sorts, sure. And so there's a fridge in the power supply. Yeah, exactly. And and it should be so once you do it with one object, you would like to take it out, enjoy and then put a new one inside the same fridge.
Ultimately, a particular fridge, simply because it's imperfect, will at some point break. But the point is that you can build a better one that lasts longer. And as far as we know, there is no limitation to how well you can approximate the perfect fridge, considering also the power supply that comes with it. Whereas for other things like copiers for quantum states, for states that are not orthogonal in quantum theory, so that this perfect measure for, say, position and velocity of an electron
It's not that any actual instance of this measure is imperfect. It's really that you cannot construct a machine of that kind. So there is a limitation to the accuracy that this measure can work. And this accuracy simply cannot be increased beyond a certain value. And this is a law of physics that's built into quantum theory. So these are the two different statements.
One statement is any particular instance of a real constructor, a real approximate constructor will at some point break down, but we can perform, we can build a better one. And that's the case when the task is possible. When the task is impossible, like in the case of measuring position and velocity of a particle, of a quantum particle, a measure for those two
simply cannot exist, which means that you can build very poor measures of those two things that will be wrong most of the times when they're trying to measure both position of momentum, position of velocity, and they cannot improve beyond the certain accuracy. There's like a finite limit beyond which they can't go. And these are two very different situations.
Do you think of constructor theory as a law of physics or as a framework for explaining physics? Like, is it more general than just physics? Is it a paradigm in science rather than a theory of science? Is it a way of going about investigating? I'd say it's both in the sense that so it has some laws of its own that are formulated in the way that I said. So stating what transformations are possible and impossible. But also
because it's somehow phrasing this different way from the standard traditional dynamical law plus initial condition type of approach is also a new paradigm. So it's not just, um, as a physical theory with new laws in it, it also has, uh, the value of being a new framework or a new language to express laws of physics. So it has both components, but the most important one to me is I think, um,
Well, they're both important, but somehow the one that that intrigues me the most is the fact that it should ultimately allow us to say more than if it pans out as we expect more than what the current laws say. So it has a physical content of its own, which is non-trivial. Otherwise it would be just a framework to rephrase things that we already know, which could also be interesting, but somehow perhaps is less interesting than say this, um,
You hope to have as an emergent theory, quantum theory and general relativity. Yes. So yeah, so I, so another aspect of constructed theory is that perhaps this is a compliment supplement to the previous question you asked. So the,
The way in which constructive theory works is that it doesn't pin down with its principles one specific dynamical theory. So we are hoping that, and I think we demonstrate this for quantum theory at least, and we're in the process of doing this for general relativity, that both quantum theory and general relativity are compatible with constructive theory's principles. So for instance, if you take the principles about information, we know they are compatible with quantum theory.
and we also have arguments to say that they are compatible with general relativity. And in that sense, they are nice because even if you are maybe skeptical about the fact that these principles are really fundamental laws, so you can be agnostic about whether this is a better way to formulate the laws of physics, you can still find them useful because you can still find these principles of constructor theory very useful because if they are things that are obeyed by both
Quantum theory and general activity that we know don't, um, you know, don't go together as, as theories themselves, because general activity is a classical theory. It doesn't have quantum effects, whereas quantum theory is quantum. Uh, you can appeal to these more general principles of constructed theories to provide explanations and make predictions in a context where both general activity and quantum theory apply, but we don't know how to put the two theories together.
But then we can appeal to these more general principles that do apply in that regime. And that's very much of interest for testing quantum gravity, because that's exactly the regime where we know that maybe one of the quantum theories of gravity that have been proposed may apply, but we don't know which one is the right one. And so having these principles are more general is very useful because you can appeal to them. And the fact that they work both for GR, for general relativity and quantum theory is a plus.
I'm thinking of quantum theory also as a bridge. These principles could be useful to guess a theory that goes beyond quantum theory and GR.
as well as to help us find predictions or experiments that can test the realm where this theory is relevant. So this realm with, for example, testing quantum gravity effects is, well, it's one of the applications of these principles of constructive theory. And that's the stuff I was mentioning earlier I've done with Vlatko. So thinking of constructor theory as some high energy theory that in the low energy limit reproduces the standard model or GR
is the wrong way of thinking about it.
Yes, I think that's the standard way of going about some theory of everything. That's actually a great thing you said there, because that's the standard way to go about this thing of finding quantum gravity theories, right? So you think, well, you know, I've got these two things, you know, I've got quantum theory, no relativistic quantum theory, and then I have relativistic theories like either GR, general relativity or special relativity.
Then let's find a way to put them together mathematically, right? And then there are proposals how to do that. We have quantum field theory. Then we have some quantum theories of gravity that can work in a low energy regime. Then there are those that actually work at higher energy, et cetera. And each of those will give you a prediction.
Unfortunately, most of the predictions can be tested for the quantum gravity predictions are very difficult to test. And then so what you do is that you say, Okay, well, let's just look at some regimes that are experimentally accessible. So that's the kind of logic that you have when you present those theories. And you're hoping that one of them will be, you will find an experiment that corroborates one of those quantum theories of gravity and refutes
effectively the classical theory of gravity, which is general relativity. So you would like ultimately some sign of the fact that gravity is quantum. So it doesn't obey general relativity after all, because general relativity is a classical theory. Now, in constructive theories, we are taking a different approach. We are considering the set of, if you like, symmetries or constraints that
both quantum theory and general activity satisfied together. So it's an exercise of saying, okay, I don't really look at the specific dynamical laws that these two theories have. But I'm trying to extract some deeper symmetry that they both agree on. For instance, the fact that both allow for things like observables. There is a concept of an observable both in quantum theory as well as in general activity.
And these observables obey certain laws of information theory and constructively we can express those laws. And then by looking at this common area of, let's say, agreement of the two theories, you can consider this deeper structure where the two theories agree, which requires you to
forget about most of the formal details of the two theories. So you will be throwing away various aspects of both quantum theory and general relativity. There are specific formalisms, but you're looking at this deeper structure where they do agree. And for example, the information theoretic structure that concerns observables is something that the two theories agree on. So the fact that there are local observables that you can measure some of these observables and so on is something that the two theories agree on.
Then, of course, general relativity is classical, quantum theory isn't, but I think there is a fundamental structure of observables that they share. They also share things like locality and other features of information theory that are in common. And you use these things, these common constraints, to make some prediction about these regimes where
quantum system is in interaction with gravity and this actually is very powerful because it allows you to imagine experiments that are in the low energy regime but they allow you to extract quantum features of gravity in a way that's easier than ways that somehow were proposed before to test specific quantum theory of gravity and this is very exciting because it tells us that
Quantum effects in gravity can actually be easier to capture than it was previously thought. And it's nice that these tests that I'm discussing as part of my work with LAPCA and other collaborators, these tests are really somehow probing a regime where we're not going to very high energies, so it's easier to actually access those regimes.
And they rest on these general principles of information theory. They're very robust and they obey both by general relativity and by quantum theory. So that's maybe the way in which it's nice to think about constructive theory as being relevant to this problem that you mentioned. And you mentioned earlier post quantum, the word post quantum, which rings a bell to me with Jonathan Oppenheim's stochastic gravity. And I believe you're both in the same university, so maybe the same departments. Is that the case?
I think he's at UCL. We are in the same kind of, but we are interested in the same topic to some degree. I guess quantum foundations, broadly speaking. I think he's in UCL and I'm in Oxford, but it doesn't matter. It's no, no, no, it's, it's, it's true that, that we have some common interests. Which is British accent. Yeah.
And yeah, I think, yeah, I think approximating a British accents is very difficult for me, but I find my best sort of, you know, blending. But but yeah, I think the, the thing that you mentioned is very relevant because, okay, that's an example of a theory that is specifically going to describe gravity and quantum systems in a way that gravity is classical.
So that's so, you know, broadly speaking, we can divide physicists into two camps. One camp says that gravity is actually quantum and we only, you know, we just have to find the right set of conditions to show that it is quantum, uh, with an experiment. And also we have some, uh, ready theoretical proposals for quantum gravity, which exists. Um, some of them are, um,
been developed in the past decades and they sort of have some predictions etc. But let's say no matter which particular proposal you favor, I think if you are in this camp your heart is saying that gravity is quantum mechanical after all, whereas general relativity says it's classical. And then there is a different camp that says that gravity is actually classical.
And when it couples to quantum system, various things can occur, but ultimately will cause the quantum system to derail and become somewhat less of a quantum system. So it will become more classical than it should be.
In this camp, you will find not just Jonathan Oppenheim, but many people that have been proposing for four years these dynamical collapse theories. Gerardi, Rimini, Weber, there are all sorts of big names there. Roger Penrose with his collapsed wave function due to gravity. So there are all sorts of great minds have been sort of powering this camp in many different ways.
Which one do you belong to? I belong to the former, so I think gravity is not just, I think gravity is quantum mechanical and I think we just have to wait for the right experiment to be performed. But the exciting thing is that this theory that you mentioned that Jonathan Oppenheim has put forward, together with many other theories of classical gravity that have been proposed over the years,
can be refuted by this experiment that I was mentioning earlier. And this is what makes this experiment very exciting because it can allow us to find a spot where we can at least tell whether gravity has some quantum features or not. So the test would entail two quantum objects, two masses being
quantum correlated entangled through gravity. So if gravity is capable of creating these correlations between two masses, then we can use this argument from constructed theory to conclude that this actually the gravity has to have some quantum features. And this is nice because it will allow us to rule out immediately all classical theories of gravity, not just general relativity, but also other proposals like
those I mentioned earlier, the collapse theories, quantum field theory in curved space time, which is another theory where gravity is classical, Jonathan Oppenheim's classical gravity theory and many others. And I think this is very exciting because it's something that hasn't been done so far. And that's the reason why there are still these proposals to say the gravity is classical after all. And it's quite, if you think about it, is really a very important issue at the heart of physics because
In order to find a good quantum theory of gravity, you need to be motivated that gravity has to be quantum. But if, say, part of the physics community is already even doubting that gravity is quantum, then there isn't much of an incentive to look into quantum theory of gravity. So somehow it will really be important to do this experiment because it will allow us to at least say, OK, now we can give up on this idea that gravity is classical. Let's really get on with it and try to find the right quantum theory of gravity. We don't even have that confirmation.
Experimentally speaking, so it's really nice to have this kind of experiment out there and it's something that people are working on to to actually implement these days So it's really it's really an exciting thing How do we know when to take a limiting theorem like a no-go theorem at one level? So for instance, you said the no cloning theorem before and then apply that
under constructor theory when we're already thinking that it may not be the case that the laws of quantum mechanics or quantum field theory are the final laws there may be something else that's underneath it. Why are we taking what's a no-go theorem up here and applying it to something that's more fundamental?
That is a great question and I think it's part of this
Search for the more fundamental, the deeper structure of quantum theory. So. So if you take quantum theories, it is there are lots of features in it and they are all packaged into the same into the formalism. So if you think for those who are not maybe specialist, the mathematics is a very powerful language.
When you write an equation of motion like in quantum theory, it gives you lots of things just condensed in this equation without giving much depth when you're looking at all of these features. Some of them are deeper than others.
For example, you can have particular features of the mathematical formalism that you're using to express the dynamical law of quantum theories, which may be parochial in the sense that they happen to be relevant for quantum theory, but they're not really fundamental. And then there are some other features to do with the fact that the laws are local,
That they are, for example, one to one. So they they map, you know, a set of states. There are not two states are going to the same state. So, you know, you keep different states into different states. That's not that's an important it's called logical reversibility. That's one thing that comes for free in the laws of quantum theory. But it's a deeper feature of them that also shared by classical
And then there are things like what you just mentioned now, the no cloning theorem, which is a thing that you can prove mathematically from the laws of quantum theory, but have a deeper essence in the sense they are part of the of the set of constraints that power the information
Theoretic structure of quantum information or quantum systems. And so by using an information theoretic perspective, you can see that this no cloning theorem is not just a mathematical feature that happens to be true of specifically quantum theory, but it's a thing that holds promise for being a general feature, something that even if quantum theory turns out to be wrong, so you have to modify the formalism and so on.
It's very likely that this feature of not being able to copy the different states that don't belong to the same physical observable is a feature that will stay. So whatever modification you do to the mathematics of quantum theory. It's very likely that it will be or it's inevitable that it will be. Well, it's difficult to say it's inevitable because these things are
matter if you like you know as a physicist you also have a matter of taste if you like you know you're thinking of things according to your own philosophy and and so on but i would say that there's there are lots lots of good arguments to expect that this feature will stay uh simply because um it corresponds to the fact that some transformations are impossible
We know that they're impossible. We've even done a number of tests that actually indirectly test this feature. And so it's an
If you like, it's to do with the operational information theoretic structure of quantum theory rather than with the specific law of quantum theory. And so it's very natural to imagine that the next theory will conserve these features and maybe have more that are even more exotic and exciting. Just like in the case of classical physics, the fact that
So if you think of classical physics and quantum theory, the information theoretic structure of classical physics has been maintained within quantum theory, but then there are extra features as we know. So the fact that, you know, you can have an observable in classical physics is also true in quantum theory. It's just that now you have extra properties of these observables that are even more interesting or exotic. And so it's reasonable to expect
You can make some arguments. I see in fact that these features that have to do with, for example, with no cloning, that is a constraint on what you cannot do with certain information theoretic variables will stay in the next theory. And maybe more constraints will come along and perhaps more interesting properties will be there, but they want undo this change from classical physics.
So to me, it's very unnatural to think that we'll go back to the structure of classical physics where, say, you can clone any state. And so in a sense, this theory, the idea of taking some of these properties of quantum systems,
their information theoretic versions as general guidelines to describe post quantum systems is a guess, but I think it's a well-informed one. So somehow we are thinking this is how it's going to pan out. But I have to say, many people would disagree. So there's a lot of debate there. It's quite hot as a topic.
And this may be why we can't make progress in certain directions because there are two diametrically opposed ways of looking at things in these physics circles. And that's usually where the fun part of science is. Yeah, that's right. Yeah, exactly. I still want to get to some of your papers, your recent ones. You have a couple on ghost particles and how you can possibly even detect them, which I would love to know more about because it's my understanding that they're undetectable by their nature.
So we're going to get to that, but I believe in one of those papers, you said that there were inequivalent representations of QFT and that that not only has some interpretive issues for philosophers, but also for the mathematicians working on curved spaces or curved space times. So can you please explain what you mean by that? And does this pose a problem to all theories of quantum gravity, even string theory? Right. So, um,
I don't think that this is necessary. So starting from the last bit of the question, I don't think this is necessarily a problem for quantum gravity theories. So the way this is some work I've done with Vlatko, the way we wrote this paper, this couple of papers, was to investigate some foundations of quantum field theory.
and also of the theory of gravity that's called linear quantum gravity, which is a low energy approximation that on which all of the quantum gravity proposals we now converge. So string theory, loop quantum gravity, et cetera. I agree with this theory that is basically a field theory for gravity, quantum field theory for gravity in the low energy regime.
And all of this was informed by this experiment that I mentioned earlier, and by also some more constructive questions that have to do with what counts as an observable in field theory, both in quantum field theory for electromagnetism, so for the theory of light and for the theory of gravity in this low energy regime. And there were some surprising answers to this question.
And this fits into a more into a broader agenda, if you like, or a broader philosophical take that I have on things. And that I think Vladko also resonates with me on as far as some aspects are concerned, which is that so quantum field theory is a theory that has lots of issues. So it's a problematic theory and in fact was proposed by those who invented it.
more like as a sort of recipe for making calculations as an approximation in, you know, in the while waiting for a better theory. Exactly. Yeah. However, what happened during the years is that while the initial, you know, the founding fathers of quantum field theory did know that this was a sort of collection of tricks to make calculations, but didn't have strong foundations, philosophy and theoretically, the next generation of physicists, which includes
Many, and finally, also myself, we've somehow forgotten about this fact. And so we are using it to calculate all sorts of things successfully, but somehow we've lost sight of the fact that the foundations are shaky.
So this is an attempt to look into the foundations, go back to look into the foundations. And I think there are many other people who are doing this. So not everyone has this view. But I think the prevailing view is the quantum field theory is all right as it is. And I think that isn't so. And these papers were
Just as a quick point, when you say the foundations, some people who say I work on the foundations of quantum theory, they mean the foundations of quantum mechanics most of the time. They don't mean I work on the foundations of quantum field theory, which is different. Correct. Yeah, that's right. Yeah, that is true. And so quantum mechanics isn't explicitly relativistic.
So if you want to put it together, at least with special relativity, you need to modify it. And I think quantum field theory is exactly what, you know, is an attempt to do that. And the interesting thing is that even though quantum, no relativistic quantum mechanics is, from the information theory point of view, is more or less equivalent to quantum field theory,
And also it's also local, doesn't violate, doesn't allow you to signal faster than light and all of these things. It still doesn't have all the features that are satisfactory as far as special relativity at least is concerned. So that's why you have to upgrade the theory to quantum field theory.
However, when you do that, there are lots more problems that come in that occur and these problems are maybe not so important for say making predictions about particle physics and that kind of enterprise which is going well and everyone is happy with what's going on there.
To some degree, but they are still important for the foundations because they inform the way we think of the next theory of quantum gravity itself, et cetera. And so that's that's why we are dissatisfied with it. And we would like to make it, you know, to draw attention to the fact that there is a problem. Those two papers specifically. We're looking at to at so so at both gravity and
the electromagnetic case, but we focus on the electromagnetic case as an example, both because it's mathematically simpler and also it allows you to make the case in a more transparent way. And to cut a long story short, the way we think about this is that when you try to construct the quantum field theory for the electromagnetic field,
The usual procedure is that you start from the classical theory of Maxwell's equations, if you like, and then you apply what in jargon is called a quantization procedure. So you can think of it as a sort of machine that you are, you just got the handle, you put in a theory that's classical and out comes something that's quantum. So it's a, it's a procedure that's been put together by various people in a few decades ago. And
The problem with that is that you can follow different paths to perform this quantization. And even though they all agree on certain observable effects, so experimental outcomes, they are not equivalent physically speaking. So, for example, they have for the electromagnetic field, different ways of describing the quantized electromagnetic field.
have different numbers of subsystems, so that photons are the quantized element of the electromagnetic fields, and different ways of quantizing it leads to different kinds of photons being there. And there is one way of doing this quantization which is explicitly compatible with
Relativity and this is in jargon is known as choosing a given gauge which is called the Lorentz gauge and when you do this you have basically four kinds of photons that behave quantum mechanically and they are part of your electromagnetic field that is quantized according to this procedure and the typical way of thinking about these things is that
two of these four kinds of photons are only there as mathematical entities, but they're not really measurable. So they can't really do much. So you can't observe them. They are just tricks, mathematical tricks that are useful to do your calculations, but you shouldn't be able to, not only to detect them directly, so to have a click from these photons,
but also they shouldn't be detectable otherwise so they're deemed as ghosts because they are there mathematically to help you make calculations but they are not essential in fact there are different ways of quantizing the field the electromagnetic field that only have let's say two kinds of photons not four and so these two ghost modes or two ghost photons are not there yes in those other ways
And why is it okay? Simply because given that they're not observables, no one cares. And we are all happy that in some other ways of quantizing the field, they're not there. Now, if you so this is, this is the usual story. But the problem is that if you look at the particular kind of experiment that I mentioned earlier, where you have now two charges, not two masses, they interact with
the standard electrostatic quasi electrostatic sort of Colombian force potential, if you like, and they get entangled through this, through this interaction. So it's a very simple problem, you've got two charges, they're interacting with one another, and they get entangled through this interaction.
If you want a local description of what's going on, meaning a description that satisfies locality at each point of the description, you're forced into using this mode of quantization that uses four kinds of photons. And particularly the ghost photons are very important in the local dynamical description of how the entanglement comes about. So in the papers, in both papers, we make a point that
There is a way to indirectly detect these ghost photons by looking at the phases that you can create on these charge probes. And this is a thing that hasn't been thought about by people who usually deal with quantum field theory, because they usually think in terms of input-output scattering amplitudes. This is one kind of bit of jargon to just look at
They look at certain physical processes are very natural to look at in certain contexts, specifically particle and particle physics and quantum field theory in the sort of traditional way. But if you look at quantum theory from the quantum information point of view, where the emphasis is on phases and on things that you can extract out of charges once they interact with the field.
It's very, if you draw, if you use this principle of locality and various other principles to sort of guide your analysis, you will see that measuring features of the two charges among which there is the entanglement between them that's caused by this static interaction is equivalent to measuring accessing these ghost photons as dynamical degrees of freedom. And this is something that
It's very interesting because somehow it seems to at least contradict the standard way of thinking about these ghost photons. And it can also be the same argument you can carry it out in the gravitational field case in the linear quantum gravity regime. It just is more complicated. So instead of photos, you have gravitons and there are more kinds of gravitons involved.
But the idea is the same. So the idea basically is this, that if you insist on having a local account of what's going on in a very simple quantum information experiment that involves two charges or two masses or even just one charge in the field, you have to somehow come to terms with the fact that these ghost modes are indirectly observable. So you cannot measure a ghost photon in the same way that you can get a click out of a photon. Yes, yes, yes.
But they are important. There are degrees of freedom that can be indirectly uncovered by measuring features of the charges once they interact through the field. It's as if the charges got clouded with the field and then by measuring the charges, you're extracting features of the field. And unfortunately, you're extracting these ghost features and not the ones that are supposed to be measurable. And this is very interesting. I think this is a sort of we are hoping this will cause some disruption in the field also because we have some
proposal for an experimental test so that the theory that we have can be tested and we're interested to see what happens when we you know when this test can be performed and you can do it both with gravity but also with electromagnetism which is probably easier to do considering you know what we can do experimentally at present man a fantastic name for a theory would be ghost gravity you could write a book on that it's true yeah yeah yeah it's a great it's very
Yeah, it's very exciting. Yeah, that's true. OK, so what is the physical interpretation then? Because ghosts come about from gauge fixing, which is just something you do to make the math easier akin to if you care about the derivative of a function only, then your regular function can have plus any constant. It'll just go away when you take the derivative and you can set the constant to be whatever you like for whatever reasons, calculation reasons. But the constant goes away when you take the derivative. That's a great point you're making there.
So the gauge fix, so the gauge fixing is exactly what you said. So gauge is corresponds to, if you like these different ways of, okay, let me, let me, let me make a step back and try to explain this a bit more clearly. Okay. Gauges are simply different ways of, of describing the electromagnetic field.
even classically. So even classically, you have different gauges. So, and as you said, it's, they mathematically, they correspond to switching to different variables for your Maxwell for Maxwell's equations. So Maxwell's equations can be written in terms of fields, electromagnetic fields, and they have some classically speaking, but you can also change the variables to express the equations.
And so instead of the fields, you can use these things called potentials, vector and scalar potentials. And the potentials are basically just a different, you know, mathematically speaking, they're just in variables. However, there are many different changes of variables that you can make, and they're all equivalent, they all collapse to the same Maxwell's equations in terms of fields.
And each of these different change of variables that may be useful for computational purposes is called a different gauge. So each of them, each of each gauge has their own have their own names. So there is like Coulomb gauge, Lorentz gauge, scalar gauge. So they have names according to how they were discovered. Right. Now, classically speaking, this is irrelevant, meaning physically relevant in the sense that you can solve the equations with fields or with potentials in one gauge or another, and no one particularly cares about what you're doing.
However, when you quantize the field, things become different because each gauge corresponds to a different quantization procedure, if you like. And so the Lorentz gauge, which is the one that is local and explicitly Lorentz compatible, so it's explicitly compatible with special relativity, leads to these four
Modes that for ghost sorry for kinds of photos which in jargon are called modes and two of them are ghosts because if you follow different quantizations which start from different gauges not the lorenz gauge but something in some other gauge for example chrome gauge.
There are only two such modes or kinds of photons, and so they are ghosts because they ultimately are somehow usually thought of as being unphysical because they are not present in all gauges. Okay.
So gauge fixing means you pick a particular gauge and that corresponds to some mathematical constraint being there. Usually this constraint is supposed to be irrelevant physically because, as I said, from the classical point of view, all you care about is just Maxwell's equations. And for Maxwell's equations, the potentials may not even exist. And, you know, they're just expressed in terms of fields. Now, the
The significance of what we discussed, which by the way, I think was discussed by other people. There's a Bernard Kay who's a researcher at York that has also made similar comments recently and in the past other people have also made similar comments, just perhaps motivated by different reasoning. So what we said is that in the quantum case,
Unlike in the classical case, there are some kinds of experiments that you can perform on the charges, which are the things that you interact with, that you use to somehow extract features out of the field. There are some experiments that you can perform on the charges, if you think that they're quantum, which are not there in the classical case, obviously, because in classical physics, charges are also
And once you go along with this fact, and if you quantize the field and you want the whole description to be nicely local, et cetera, you are forced to see that the charges get in some situations, they simply get to depend on the degrees of freedom that are supposed to be goals in the Lorentz gauge. And by then measuring the charges in certain situations that are possible now to measure, because we have quantum metrology facilities that allow us to do that,
You are indirectly accessing these goals. Yes. So in a sense, this idea of gauging variance, which is very important for classical electromagnetism, is also relevant for quantum electromagnetism or quantum electrodynamics. But it doesn't forbid us from, it doesn't impede the realization of these experiments that we discuss in the papers.
And so it forces us to revisit the idea that the ghosts are not measurable. They are not measurable in the sense that they may not be measurable in the standard sense of being measurable. So in usually standard sense means to detect a click. So, you know, you have a photo in submitted and you detect it. This is one way of being measured, of measuring features of the quantum electromagnetic field. But there are other ways of probing it with quantum charges.
Which don't necessarily amount to detecting a photon of the ghost kind, but they amount to some kind of detection. It's not a direct detection of these photons, but they are detecting some quantum features of these ghost modes. So in a way we are suggesting that the idea of measurability and what counts as measurable should be revisited in light of the fact that we can actually perform these experiments. So would an analogy be like the Aronov-Bohm effect where before that the
Electromagnetic potential was thought to be something that was just mathematical, a convenience. And then afterward, you still don't detect the electromagnetic field directly, but you see its effect on the phases of the electrons that go through a soliton or the outside of a soliton. And that, by the way, was revolutionary. So this paper with you and Vidral, is that correct? Yeah. Let me read its title. Interference in quantum field theory detecting ghosts with phases. And I'll put a link to that in the description.
So this paper, is it suggesting something physical? We think that there's something physical about the electromagnetic field with one of the major pieces of evidence, historically being the Earnhardt-Bohm effect. So with your proposal to detect ghosts, are you saying that something else exists, like something that's choosing a gauge? Is there something else that some other field that is actually in existence that we thought was a mathematical trick?
Yes, I think this is one way of thinking about what we do. In fact, it's very related to the Arano bomb effect. In the following sense, we've also actually
thought about the Iron bomb effect and somehow this was part of the way in which we landed on this idea. I see, I see. Because even in the case of the Iron bomb effect, the interaction between the electron and the solenoid is mediated by accident.
They're very similar words, somehow different things, but the solenoid being this thing that generates a magnetic field only somehow ideally along a line and zero field elsewhere.
And in the A-B effect, you have this interesting fact that if you have an electron that goes exactly in the region where the field doesn't appear to be there, it can still be affected by the solenoid because when the solenoid is switched on, the electron has a different phase when it interferes compared to the case when the solenoid is off. So this shows that the electron is actually picking up some signal from the solenoid, even though there is no direct field that's acting on it.
Okay, there the issue was how does how do the solar and the electron communicate? Even though classically speaking, if we describe the whole thing classically, sorry, if you describe the field classically, it looks like there's no field at the point of the electron. And the key to answer that is that even though classically that's true, if you consider the quantum description of the field, you will notice that
First, there is a back action of the electron on the solenoid. This is something that was already pointed out by Lev Weidman, who anyway used still a classical model for it. And second, more importantly, I think there's the fact that even though the field is zero classically outside of the solenoid,
a quantum field can never be zero in the strict sense. And so the so it can be zero. Some aspects of the field can be zero, but there is still a quantum feature of the field that's there. And some of the photons that are so there are photons coming in a sense, going back and forth from the solenoid to the electron, vice versa. And these photons happen to be also of the same kind as those that we
the most
and this is only clear the fact that everything happens locally and nicely can only be accounted for as far as we know in this in a particular gauge. So somehow this set of results that we mentioned show us on the one hand as you said that some gauges are better than others the Lorentz gauges is
more accurate in the description of what's going on, because it's explicitly local and also Lorentz covariant. But also there's the fact that in order to have a complete description of what's going on, you have to quantize the field. So you have to have a quantum description of the field. Otherwise, as it happened in the case of Harnov and Bohm, if you stick to a completely classical description of the field and you want the charges and the solenoid
Sorry, the field is classical but the charge is quantum. You will incur into some issues of the description and Arnold and Bohm somehow were concerned that this effect, as they described it in the semi-classical theory, appeared non-local. And this can be cured by the fact that you quantize the field. So the lesson in this set of experiments is both that
Some gauges are more accurate than others physically speaking. They're more realistic. They make more sense. So they tell a more coherent story because they're all explicitly local, etc. Lorenz gauge is one of them. The second lesson is that in order to have a local description, you always need to quantize everything in your systems. Otherwise, if you insist on one of the systems being classical and the others not, you will incur some issues typically with locality. And finally, the third lesson is that we will need if we quantize the field and you look at
some of these effects in specific gauges like the RANS gauge. And you consider all the experiments you can perform on the charges, not just these input, output scattering, and amplitude things that people usually look at in particle physics, for example. You will notice that it's inevitable to have to modify your notion of measurability because even though you cannot detect
As far as we know, directly these ghost photons, by say having a click in a photo detector, you can indirectly detect their degrees of freedom by making measurements on the charges. An inescapable conclusion of how you analyze the analysis that you can make of the situation in the Lorentz gauge. And so this requires us to just simply enlarge the set of things that we can
So this, to me, sounds like a departure from constructor theory, but you see it as tied or you see there being implications. Oh, yeah, absolutely. Yeah, I think in my head, all of these things. So if you like, you know, if you look at I'm being very
Now I'm sort of telling you somehow that things do it the way I think about stuff so I don't know if it's really relevant but the way I think of this is really you know you can look at these papers they look different from each other but I think it's really me trying to understand how to apply this notion of observable that's more general that you have in constructive theory of information so this paper I wrote with David a while ago
where anything that is copyable, any set of attributes that is copyable can be considered as an observable, whether or not it has some formal features that quantum theory requires, such as being Hermitian operator, et cetera, et cetera. So it's my way of understanding how this concept of an observable can illuminate different parts of quantum field theory or quantum theory more generally.
Especially those cases where we don't have an understanding of what we can really observe. So quantum field theory is an example of a case like that because, as I said, even though it's a quantum theory of fields,
It has issues with various foundational aspects. And so what counts as an observable there is very confused. And I think the foundations of quantum field theory also are not so clear about what is an observable and what isn't. And so this is a way to, in a way, show the power of, well, I don't know if it shows the power, but let's say it's a way of applying
These constructed theory notions of observables that are more general than particular notions in quantum theory or quantum field theory to help us find new ways of probing fields compared to what we used to know before. This is a bit technical, but when I was looking up the categorical approach to constructor theory, I saw that what you want to do is you want to divide your space into what's impossible and what's possible and your spaces then or tasks. I believe tasks are the morphisms in a symmetric monoidal category.
and I believe when you divide your space into what's possible, it becomes a subcategory. Do the impossible states also form a subcategory or no? I actually don't know because the category theory can be applied, I'm not an expert at all, but I think it can be applied to a number of things. In fact, it's so flexible, it can be applied to a very broad set of theories in a way, right?
And I am not sure about the specific answer to this question. I think what is promising there, and I was very happy that some people were interested in category theory, actually decided to do this work, is that it could be that by casting constructed theory in these terms that are, if you like, more modern from the mathematical point of view as well,
I think there's a chance that some of the theorems we have could be condensed in a very elegant short proof and as well you could also see unifications between things that we think may be distant and not related. I'm really hoping that that continues and that perhaps more
More people may come along and have more results in that kind of direction. But as far as your question, I'm not so sure. The only thing is that if I sort of understand correctly what you were saying, what allows you to conclude that is the fact that in constructed theory, we have this principle that's called the composition law.
that says when you compose two possible tasks, you still should obtain a possible task. Whereas this isn't true for impossible tasks. Okay, so then it wouldn't form a subcategory. Yeah, they wouldn't. Right. So the idea is that somehow this composition law only holds for possible tasks. And it doesn't for impossible tasks. So in fact, you can have two impossible tasks that compose together give you a possible task.
A typical example is the so imagine again, the task of high of first raising the energy of a system and then lowering by and then the second task is lowering the energy by the same amount. You can compose the two tasks joint, right? So they're both impossible, right? As we said earlier, yes, yes, by themselves.
But if you compose them, you basically have the task of the identity, right? So you stay in the same place. So you go, you know, one task goes up, the other goes down by the same amount is basically the task, the resulting composition gives you the task of staying in the same place, which is a possible task by somehow axiomatic, by assumption. So that's an example of two impossible tasks giving you a possible task. So I'm guessing
Because there isn't disclosure property, presumably the answer to your question is that the impossible tasks should not have that property. But as I said, I'm not sure. And the principle of locality, that's super important to constructor theory. It's that different than locality in the regular physics sense? No, I think it's so OK. Locality is a very subtle concept and has lots of different meanings in physics.
So in other words, if you have a way to describe your, you know, you have a system that's made of subsystems,
and you have a transformation that only involves one of these subsystems. It should only be able to change the subsystem in question and not the others. And this applies both to the directly observable things on the other systems and those that are not directly locally observable. And this is a property that is true for quantum theory,
Non-relativistic quantum theory for quantum field theory is true for general relativity and is true for some forms of, I would say, well, it depends what you're saying, but I think somehow this is true for all the theories that we think are reasonable, even non-relativistic quantum field theory that somehow is not necessarily the correct theory because it's not relativistic.
Now, this is different from the concept of locality that's sometimes referred to as Bell locality, which is a notion that's very important in quantum foundations because it's the property that a theory can be described by a local hidden variable model that uses real value stochastic theories.
And John Bell had this very interesting result, which was a mathematical theorem that says that if you violate with some statistics, certain inequalities, which are called Bell inequalities, you can conclude that your statistics are coming from a theory that is not Bell local, so it cannot be reduced to a local hidden variable model.
And quantum theory is not Bell local, so it's not describable as a local hidden variable theory that uses real value stochastic theories. So in that sense, it's often confused. It's often said that quantum theory is non-local, but what one means there is Bell non-local. So it's the fact that quantum systems can violate Bell inequalities. But the locality we're talking about is really more basic or fundamental. So it's to do with the fact that there is no action at a distance and
Even quantum systems that are entangled, that violate Bell inequalities, do not allow you to perform this action at a distance. And so that's the feature that constructive theory cares about. And I think this locality is really built in in the foundations of the theory. It's a very important axiom, if you like, or principle. And it's also very important in the context of these experiments that I mentioned earlier, the ones to test quantum gravity and also the
Electromagnetic discussion we had early with the goals and also the gravity version of those results. So yeah, it's a very important principle. Chiara, what's a no design law, a quote unquote no design law? A no design law is a law that is very important for understanding this is in the origin of life problem.
Frame of, you know, frame of mind. So we're thinking in that camp. Um, so the problem of biology is to one of the problems of evolutionary biology is to explain how, uh, living systems can have come about without, um, that being a designer. So typically I think, you know, Darwin was battling with the idea that somehow there was a
believe that you had to appeal to a creation sort of, you know, to God being there to design to a designer to create entities are highly complex out of simple beginnings. And so if you don't want to go that way, because you don't want to make that step to believe in such an entity, you're in luck because Darwin and
All the great biologists who came after him and kind of refined his idea showed us that it's possible to have a complex entity come out to simple beginnings, simple initial conditions that don't include complexity.
simply by waiting for long enough and allowing for these natural selection and basically mutations that are not specific to what you want to get in the end. So you need the mechanism of natural selection to be there and you need the possibility of there being things that can replicate which
you can call genes in a broad sense, a bit like what you know, with the Dawkins terminology. And so they're not just genetic, they're pieces of, you know, strings of information that can replicate in a stable manner. And then you need the possibility of having errors in these replications that introduce variations that can then be selected by the national selection occurring in a given environment.
Now this, this concept is very, uh, this, this description is usually in biology is given at a high level. So no one really thinks of what about the laws of physics? Um, we, you know, people just think, okay, the laws of physics are in the background. We don't care in biology. That's fine. But somehow when you are trying to imagine a way to reproduce this thing in a laboratory or in a simulator or in a computer or in some sort of dynamical system,
You have to have a very good understanding of what dynamical interactions you are happy to allow in your simulator to mimic the fact that like in physics, these laws, these dynamical interactions don't have the design of the life that you want to emerge out of your process of evolution.
And this is a very tricky concept. It's tricky even in physics itself and in simulation is even more tricky. But the point of no design laws, the definition of it is really to highlight the importance of the fact that when you're running an argument like Darwin's, you want to make sure that you haven't snuck in some assumptions about the laws that are
you know, that govern the interaction between microscopic entities in your biological system, which ultimately contain the design of the living system that you create at the end of the natural selection process. So a no design law is a law that isn't specifically designed or crafted to bring about a specific complex entity or a set of specific complex entities. And we know that the laws of physics says
We believe the laws of physics we know kind of govern the universe. Our current guesses are no design laws in this way because they don't specifically contain symmetries that are especially
suited for the emergence of a specific form of life. And that's very important to know, because that gives strength to Darwin's argument, if you like. And so if you are concerned with the possibility that there could be a skeptic that might not believe in Darwin's explanation,
It's good to remind ourselves that the laws of physics that Darwin's presupposes for his own reasoning and neo-Darwinism in general presupposes are no design laws. So nothing is assumed of the dynamical interactions that are used in the sort of Darwin's theory of evolution. And when you then try to, if you're thinking of a way to
to reproduce this thing in a laboratory. So you think of having or in a computer, it's very important that when you, you know, if you start with simple beginnings and you get complex entities at the end in your simulation, you want to have a criterion and a test or a check away to check that the dynamical interactions that you use in order to show this progression do not already contain the design of what you're getting at the end. So you start with something simple and outcomes, something like an elephant.
Let's say in a simulation in a computer, you want to make sure that the computer doesn't have some rules of interaction between elementary cells that somehow contains the design of this entity that comes at the end. So you don't want to assume what you're trying to prove. That's right. You don't want there to be circularity. I see. Yeah, that's right. So if you were to assume that then you would be circular because it would be like saying, OK, well, I gave you an elephant, but actually I snuck in the program, you know, the design of it. So that's not very surprising.
What's surprising and what's cool in what happened in the biosphere on earth and perhaps elsewhere in the universe we don't know is that as far as we can tell because given that laws are not designed specifically for life, life came about and we have lots of complexity now going around the earth and that's very interesting and somehow
Okay, great, great, great. Now that we've talked about design law or no design law, what is super information? So super information is a
Particular kind of, in fact, I should talk about super information media. That's the thing. So it's a generalization of the concept of quantum systems. So it's in the context of the constructed theory of information and the super information media are physical systems that obey the laws of constructed theory.
and have extra properties compared to physical systems like a bit that can only contain classical information. And these extra properties have to do with the fact that not all of the states are copyable, just like quantum systems.
And in our paper with David, we prove that these systems, the super information media have all the qualitative properties of quantum systems. So if you like, you can think of the theories that describe these super information media as a generalization of quantum theory. And a super information medium could be a medium, a physical system that can perform quantum computation.
But may not obey fully quantum theory. And so they are they are like post quantum systems can be described by post quantum theories that obey principles like locality and the interoperability of information. So all of these principles construct the theory. And these super information media are very useful. The theory of these things is very useful in the context that I mentioned earlier, where you have a quantum system interacting with something that like gravity that may or may not be quantum, because a quantum theory of gravity may describe
I see, I see. Does constructor theory have anything to say about dark energy or dark matter? Okay, this is very speculative. I think we we've remarked a number of times. This is like I said, the speculative level that so these systems are things that may or may not obey quantum theory as we know it. And
also may require some upgrade of the theories that we currently have in general, even at the kind of mathematical level. And because of that, I think it's very useful to apply to them this theory of super information media and or the constructed theory of information more generally, because they may still obey the principles of information.
and also the principles of information theory that we laid out in our paper put constraints on them. So, for example, the principle of interoperability of information which says that any system that can contain information should also be able to interact with another system that can likewise contain information and you should be able to set up some interactions between them that allow for copying various states. I think this is the thing that can be so this principle
could be applied to dark matter, for example, and it may turn out that this principle rules out some of the theories about dark matter simply because they have, they somehow violate the principle and they say that some copy-like operations between the dark sector and the non-dark sector are impossible.
So in a sense, I think constructively, you can say a lot about that. And maybe this is something for the future application, but it's definitely something that is on the, you know, something that we may at some point address. Now, given that you're working on what is like a theory of everything, or maybe it's a framework for a toad rather than a toad, but regardless, do you believe a toad to be possible? There are two respects in which I mean that, like, does it exist? So that's one question. And then even if it exists, is it knowable to us?
So I think, I don't think O'Constructor theory is a theory of everything for the reason that, as I said, it has the ambition to express. So as I said, it has a principle, it has a number of principles that can express some aspects of physics, ideally all of them, but also it may not
have the features of the dynamical theory. So, so you may not actually directly respond to the standard paradigm of the theory of everything. I think a theory of everything perhaps is not the most fruitful way of thinking about stuff in the sense that it is my philosophical position that we, you know, in any theory, no matter how complete it looks, we
may be able to find problems. And so these problems will lead to something else. And so the way I think about stuff is more like, I like to think of there being different levels of explanations. Maybe there are infinitely many, and we just keep going from one to another by understanding things more and more. So in that sense, I like this more open-ended way of thinking about science, physics specifically.
Where we just keep digging and sometimes when we dig deeper, we find a different level of explanation that brings some more unification along and we go from one level to another. And I think the reason ultimately, so there isn't really a very strong, I think it's very difficult to convince someone who thinks otherwise that this is the case, but I think I can say that the reason why I like this view is that
I think it gives you really, as a physicist or as a scientist, it gives you the idea that there'll always be something to work on. And I think that's a fun thing to entertain in your head. And in a sense, the theory of everything somehow suggests this sense of closure that you might achieve at some point. And I think given that we can never know whether what we know is true,
It seems to me to clash with this epistemological stance that I have that I think it's impossible to actually know whether what you know is really true. All you can hope for is somehow to be able to find problems in what you know and somehow change your view accordingly. So in a way, I think I
I kind of like this stance on things. I think this is very popular and if you like and I sort of subscribe to that I really like this approach and it's very scientific. I think this how scientists behave more or less even though when they you know, even though sometimes they don't acknowledge that but it seems to me this is kind of very natural way of thinking things. What would be the difference between information and knowledge? So knowledge is a kind of information that
has extra properties. So information is really some set of states that can be copied to arbitrary high accuracy, whereas knowledge is this extra feature of being information. So a set of states that can be copied, which also are capable of causing transformations to occur on physical systems and to stay
Is it possible for there to be information that isn't physically realizable like that isn't instantiated in something physically?
I mean, that's at least the way we think about stuff. That's that's very much in line with with the I think there's a long tradition of people, physicists have realized this, I think Charlie Bennett being one of them. And I think the you know, most of the founding fathers of quantum information had this bias of thinking that computers are physical entities, they run on physical laws, and therefore they have properties that are
The idea of information as being an abstract entity that's not embodied in physical systems is really something that somehow doesn't belong to the sort of philosophical viewpoint that we are following.
Um, and ultimately the construct of your information is precisely says that. So it gives you a handle on information that's physical. It says it's a set of states that can be copied. Um, and so that's, you know, when you talk about the set of states of a physical system in dark, you're saying that everything that can, you know, information is actually physical. Okay. Now, Kiara, before we go, I just want to know what's one
Mistake. What's the best mistake that you've made that has turned out well? Well, I think I think I've made lots of mistakes, but perhaps the the Okay, I think that the one that comes to mind now is this fact that I
So I think I, I didn't know, I, I, I thought for a long time that I wouldn't be interested in necessarily in, in, in science or physics. So I think my love for physics was, uh, was, came in late in my, in my life. So later, later, let's say then, then say teenage years. Right. So I think I, um, initially thought I would be a writer. So I'm about 10 or something. I had this idea that I would love being a writer and
I loved languages, I loved literature, poems, everything that had to do with language just fascinated me. And so I ended up sort of working a lot on these things and my own choice in school in back in Italy, I think where you can choose between, it's a bit different from the anglosphere, but I think you have a choice between two kinds of sort of main parts. And one of them is more science inspired, the other one is more
Classics and literature inspired. So I chose that. And okay, you could think of that as a mistake in a sense that later on, I realized that I really did want to study science. So that happened at the end of the five years of secondary school. But so on the one hand, this meant that I managed to actually just satisfy this love for literature and classics and languages.
And then it also meant that when I switched to science at the university, when I switched to physics, I actually fell in love in the second time in a sense. And that was very exciting. So unlike maybe my other colleagues that somehow had read mostly, you know, they already had had this school mostly within science. So they had being exposed more to say physics and calculus, et cetera. I really just found it then when I was 18 and
I just, I just found that it was mind blowing and very beautiful. And somehow the fact that I had all of this baggage from the classics side of things made it even more exciting for me to see these things. Uh, because I, it's like seeing them with different eyes and I, so in a sense that was maybe you could classify it as a sort of mistake initially. Uh, maybe I could have immediately started on the parcels of already back in the teenage years.
But I didn't and I'm very happy that I didn't because I somehow it's really nice to be able to see maybe similarities also between different, you know, between humanities and science and physics and physics is really a storytelling. So I ultimately satisfied my life for my love for storytelling, which is actually what I really loved as a child.
by ending up in this field that is really about telling stories about the universe and it's the most accurate, most complete way of telling stories because it really has, you know, you have reality as a checkpoint and you have to sort of make stories are compelling and clear and precise and that's just what makes me fascinated about physics itself.
Yeah, aspect of your book to bring this back that I enjoyed is that and again, the book is the science of can and can't and it's on screen. It's in the description is that you get the sense of a love of writing. Something I'm curious about is many scientists see their exposition. They're explaining to the audience as some necessary evil they have to do at the end in order to get grants.
Because they need to drum up some fervor in the public sphere, but they don't see it as something that they want to do. They like to do the research and speak technically. When you were writing your book, did you find that the writing actually helped you develop your more technical ideas? With TD Early Pay, you get your paycheck up to two business days early, which means you can grab last second movie tickets in 5D Premium Ultra with popcorn.
Absolutely, I think so.
I do find it challenging. So in a sense, maybe, you know, if I followed my least resistance path, I would also sort of tend to just do my own, you know, technique, technical bits of writing and research. But the task of explaining it to someone who doesn't maybe have the same mathematical tools and doesn't maybe even have an interest in this stuff and making it interesting to them.
is a very humbling and at the same time exciting enterprise and I think I really enjoyed it because it clarified in my head also my own understanding of things.
It didn't perhaps lead to new discoveries in the sense, in a strict sense, but I think it's true. I agree with, I think many physicists have said this before, that somehow if you only know a formula and maybe you understand the mathematics, but you can't explain it in simple terms, it means that something is escaping you and you don't understand it yourself. So somehow trying to explain it to someone in simple terms is really a great exercise for everyone to do.
Data for teaching, I think teaching is a similar thing. It's a very nice activity because it allows you to really try to sort of break down your understanding of something very complex into simple steps that you can then explain to people who haven't seen this before. So yeah, definitely, I really enjoyed it. And I really enjoyed writing anyway. So because that's, as I said, this is part of my passions.
And my last question, I know you got to get going. Where do you get your best ideas? When and where do you get your best ideas? Oh, that's very it varies. I think it's really like it's really true that you have to be it's really a creative act. I think it's something that this probably is true for everyone is doing some creative work. You have to be relaxed.
You have to be free of worries if possible. So somehow you have to be able to Bring yourself to a sub space where you're not concerned about, you know daily problems
And you have to be, at least as far as I'm concerned, you have to feel free to explore things without constraints. So I was very lucky in my PhD or the field because all my supervisors, I mean, David was a collaborator. He's great at this, but I think also R2 Records was my supervisor at the time. They are masters of this freedom seeking attitude. So I think I was
really inspired to be just so I was let let you know left on my own and to just think and I think that's the best state you can be in so you're not forced by someone to say okay you have to work on this problem or this other problem you have to be free to let your mind roam and and and explore things like you are you know as Newton used to say right that you're on a seashore and you're sort of
looking at pebbles and you're like a child playing with things. You have to be in this fun seeking free state to have your best ideas. That's as far as I can tell. And this is, as I said, a constant also, actually, my mentor, Mario Rossetti, who was the person, by the way, who introduced me to quantum information back in my masters in Torino. I think all of them, all of my mentors had this feature.
And they all said that their own mentors were like that. So I think somehow it must be a constant. It's true of, I would say it's true of most creative activities that you need to be in that state to be inspired. And you have to be able to sit quietly somewhere to be inspired. So it could be that you are on a journey. So sometimes I enjoy thinking when I'm traveling, it doesn't have to be that you're in a super quiet place, but somehow you're tranquil. And so, you know, you're in your own zone and then you can think.
Um, and, and so the, the place can vary. It could be on a walk. It could be on a swim. It could be while I'm traveling, uh, or I'm sitting at the desk, but there has to be a moment where you're sitting there in this space and waiting for inspiration. And sometimes it doesn't come, you just have to be patient and wait for it. But when it comes, then it's really nice to follow it. And, and yeah, it's, it's really very much like an artist. That's, that's, I think the way I work and presumably all my colleagues also do.
You mentioned you're waiting for the inspiration or just the inspiration occurs to you? Like, are you sitting here like, come on, inspiration, come on. Yeah, I think I think sometimes you have to be patient and wait is a bit like when you're sitting to, you know, if you're in outside walking and you want to see some rare animal in the wood, you just have to be patient. But you've got to go there.
And I think that's the thought that's maybe the hard thing to do. You have to cultivate these things, especially if you're not, you know, there are also things that go in the way I think of research these days, not just daily problems with, you know, everyday life, but I think also, you know, grant applications, as you said, and admin duties, etc. They're all enemies of creativity, optimizing research oriented ideas. So I think you really have to find
some way to guard your time and say also no to things and find some free time to wait for inspiration. And you've got to be somewhere for it to visit. So in that sense, yeah, I think I very much think of me myself as sort of being one of those explorers that are waiting for to see a rare animal in the forest. I think that's how I think of myself.
Thank you, Chiara. Thank you for spending so much time with me and the audience. Thanks to you. Yeah, it was great. Great being here. Thank you very much for the questions and for listening. Yeah. Thank you. Take care. Bye. Bye.
The podcast is now concluded. Thank you for watching. If you haven't subscribed or clicked that like button, now would be a great time to do so as each subscribe and like helps YouTube push this content to more people. You should also know that there's a remarkably active Discord and subreddit for theories of everything where people explicate toes, disagree respectfully about theories and build as a community our own toes.
Links to both are in the description. Also, I recently found out that external links count plenty toward the algorithm, which means that when you share on Twitter, on Facebook, on Reddit, etc., it shows YouTube that people are talking about this outside of YouTube, which in turn greatly aids the distribution on YouTube as well.
Last but not least, you should know that this podcast is on iTunes, it's on Spotify, it's on every one of the audio platforms. Just type in theories of everything and you'll find it. Often I gain from re-watching lectures and podcasts and I read that in the comments, hey, toll listeners also gain from replaying. So how about instead re-listening on those platforms?
iTunes, Spotify, Google Podcasts, whichever podcast catcher you use. If you'd like to support more conversations like this, then do consider visiting patreon.com slash Kurt Jaimungal and donating with whatever you like. Again, it's support from the sponsors and you that allow me to work on toe full time. You get early access to ad free audio episodes there as well. For instance, this episode was released a few days earlier. Every dollar helps far more than you think. Either way, your viewership is generosity enough.
▶ View Full JSON Data (Word-Level Timestamps)
{
"source": "transcribe.metaboat.io",
"workspace_id": "AXs1igz",
"job_seq": 6655,
"audio_duration_seconds": 7586.65,
"completed_at": "2025-12-01T00:25:01Z",
"segments": [
{
"end_time": 24.258,
"index": 0,
"start_time": 0.009,
"text": " What is Constructor Theory and how is it different now than in its original formulation by David Deutsch? So Constructor Theory is a new way of formulating the laws of physics and this was originally proposed by David I think back in about 2011, something like that."
},
{
"end_time": 54.633,
"index": 1,
"start_time": 24.77,
"text": " as a new mode of explanation. So he wrote a paper that had a very strong philosophical component which laid the foundations of the theory in the form of a program in a sense. And the key idea there was to modify the way we formulate laws of physics and so the fundamental laws of physics. So instead of using things like"
},
{
"end_time": 81.834,
"index": 2,
"start_time": 55.367,
"text": " Dynamical equations, so laws of motion and initial conditions, which is what most fundamental theories do. Switch to a different mode where the basic fundamental statements are constraints about which transformations can be performed and which transformations cannot be performed and why. So what is possible and what is impossible?"
},
{
"end_time": 96.493,
"index": 3,
"start_time": 82.278,
"text": " And then consider dynamics and initial conditions as kind of emergent consequences of these principles. So it's really like a switch of perspective into thinking what is what is the fundamental element in a physical theory."
},
{
"end_time": 120.794,
"index": 4,
"start_time": 97.21,
"text": " And this was there. I think that's the key idea. And David was really inspired to do this by the quantum theory of computation, which is a theory that he himself pioneered in the eighties when he with other people proposed the idea of a universal quantum computer there. I think it's really important to in that theory."
},
{
"end_time": 144.77,
"index": 5,
"start_time": 121.561,
"text": " To think of what can be performed by a universal Turing machine and what cannot be performed by it under given laws of physics. So in the case of classical physics, you have a classical universal Turing machine that can do certain things and not others. And then you have a quantum universal Turing machine, which uses the laws of quantum theory, not classical physics."
},
{
"end_time": 173.797,
"index": 6,
"start_time": 145.452,
"text": " And there you have new different modes of computation available. And this was a key insight in developing this idea of the universal quantum computer and construct a theory. One way to see it, and this was already there in David's paper, is to think of it as a theory of a more general programmable machine that is even more general than a universal computer. And this is a universal constructor."
},
{
"end_time": 196.578,
"index": 7,
"start_time": 174.65,
"text": " So a constructor is an entity that can be programmed to perform a number of tasks that are not necessarily computations. So you can think of a heat engine as a constructor, if you like. A 3D printer is a constructor. Anything that can be programmed to perform a given physical transformation, you can think of it as a programmable constructor."
},
{
"end_time": 224.548,
"index": 8,
"start_time": 197.551,
"text": " It just has to have this property of being able to do the transformation once and then keep its ability to do it again. So that's the key feature of the constructor, which makes it different from a different system that can just simply perform the transformation once and then maybe be destroyed or whatever. And a universal constructor is the most general programmable machine that we can think of. And this is what the physicist John von Neumann"
},
{
"end_time": 246.169,
"index": 9,
"start_time": 225.162,
"text": " Thought of when he was imagining the ultimate, you know, the most general program machine that could be built by by by humans in a sense. And constructed theory can be thought of as a way to generalize the quantum theory of computation to cover these machines that are more general than computers."
},
{
"end_time": 269.735,
"index": 10,
"start_time": 246.92,
"text": " And this is somehow a completion of what phenomenon had in mind, because phenomenon had this idea of the universal constructor, but then never really deliver the physical theory of these machines. Whereas we are hoping with constructed theory that we will be able to deliver a theory of these machines. At the same time, also deepening our understanding of physical theories, because"
},
{
"end_time": 296.118,
"index": 11,
"start_time": 270.35,
"text": " When you understand what are the fundamental limits of the universal constructor, so what is it that it can or cannot perform, you've also expressed what are the possible and impossible tasks according to the most fundamental laws of physics. So in a sense, studying the universal constructor and studying what is possible and impossible under the laws of physics is the same. And this is a key insight in David's paper."
},
{
"end_time": 323.626,
"index": 12,
"start_time": 297.227,
"text": " There are lots of other things in that paper, I think, different ways of thinking about constructive theory as a way to expand on complexity theory and chemistry and thermodynamics and so on. But back then and I, well, a few years after, I think later, something like 2012 or something, I was doing my PhD. Back then, we didn't have"
},
{
"end_time": 352.312,
"index": 13,
"start_time": 324.77,
"text": " Any specific application of the theory so it was more like a program right and what happened between then and now and let's say what I was really interested in when I started working with David and then kind of develop various things on my own was that I I like this idea this new switch this this sort of switch of perspective and I thought it was very promising and then I wanted to find some specific problems that this approach could be applied to"
},
{
"end_time": 376.578,
"index": 14,
"start_time": 353.626,
"text": " So then I think in partly my thesis and then later on in my research work, I did a few things where I applied this theory to various problems. So initially with David, we applied it to information theory and we found a very interesting way of expressing with this language of constructed theory the"
},
{
"end_time": 406.834,
"index": 15,
"start_time": 377.654,
"text": " laws, the principles that underlie physical theories of information. So it's this theory that we developed together, David and I was to express what are the regularities in nature that are needed for information to exist and also for quantum information to exist. So these are ways of handling both quantum and classical information in the same theoretical framework. And this is very important for direction of research that I'm really keen on"
},
{
"end_time": 429.667,
"index": 16,
"start_time": 407.312,
"text": " Nowadays, which is the direction where you're thinking of systems that are in interaction with objects that obey quantum theory, but may not themselves be quantum. Maybe they behave according to a new theory. Maybe they behave according to a post quantum theory. Sure. For example, gravity is one of these objects because"
},
{
"end_time": 451.442,
"index": 17,
"start_time": 430.077,
"text": " We don't know. We have various proposals for quantum gravity, but we don't know which particular quantum theory of gravity is the correct one yet. And so in that case, it's very important to have a framework, a theoretical framework to handle the situation where gravity that may or may not be quantum interacts with the quantum object."
},
{
"end_time": 474.582,
"index": 18,
"start_time": 451.988,
"text": " and a framework that can handle both quantum and classical things, let's say, in the same unified scenario. And that's what the Constructor Theory information can do for you, among other things. So it's one direction. Another direction where we made progress was thermodynamics. So there was an application of Constructor Theory to thermodynamics."
},
{
"end_time": 502.449,
"index": 19,
"start_time": 475.111,
"text": " and to expanding on the current formulation of the second law, something that we can discuss later. Right. And then a third direction, broad direction was the application of constructed theory to the physics of life. So there are these issues about how what is, you know, what is the simplest entity that can occur in the universe, which can be considered as alive."
},
{
"end_time": 529.394,
"index": 20,
"start_time": 502.978,
"text": " What are the, let's say, essential features of this entity? So does it have to be programmable in some way? Is it a kind of programmable constructor? What's the minimal structure of this entity? And in that direction, I think I applied Constructor Theory to tell us under what are, let's say, the necessary and sufficient conditions for"
},
{
"end_time": 557.022,
"index": 21,
"start_time": 529.974,
"text": " an entity to be capable of self reproducing very accurately. So just like living things do. So in a sense, you know, when we think of self reproducing entities, we think of laws of biology, but ultimately what we can do in biology is really set by the laws of physics that we have available. So it's interesting from the physics point of view, and especially from the construct aesthetic point of view to ask,"
},
{
"end_time": 578.848,
"index": 22,
"start_time": 557.585,
"text": " Considering the laws of physics as we know them, what are the minimal features that are both necessary and sufficient for a living system to be capable of self-reproducing accurately? And this is the kind of stuff that Constructor Theory can deliver on. I think I developed this branch of Constructor Theory."
},
{
"end_time": 607.756,
"index": 23,
"start_time": 579.309,
"text": " We the view of applying this to the study of, you know, for example, the origin of life and possibly the study of life elsewhere in the universe. So these are, let's say, the three macro directions in which progress on and then David has also worked independently on on there is other things to do with the universal constructor itself. And finally, the there are a few things in the pipeline with"
},
{
"end_time": 636.408,
"index": 24,
"start_time": 608.37,
"text": " Some collaborators of mine, Maria Violares, who is a PhD student, a default student here in Oxford, who's developed some interesting results about irreversibility. So again, about thermodynamics and then some work that David and I are doing on the constructed theory of time. So this is kind of forthcoming. And then some extra work on the applications of constructed theory to this area where"
},
{
"end_time": 666.271,
"index": 25,
"start_time": 636.852,
"text": " We have a quantum system interacting with something that may or may not be quantum. And this is something I'm doing with Giuseppe Di Pietro, who is another PhD student here in Oxford. So there's a lot of there's been a lot of work. And finally, there's also been an interesting application of constructed theory to the problem of testing quantum effects in gravity, which is something I've developed with Vlad Kovetrov, who's a physicist here in the physics department. So I think that's an overview of what's going on."
},
{
"end_time": 693.712,
"index": 26,
"start_time": 666.834,
"text": " Wonderful overview. Thank you so much. Thank you. The audience should know that there's a book that you have called The Science of Can and Can't, which goes over these topics in an extremely introductory manner to people who don't know even what a Turing machine is. And so I have read your recent papers and that book as well. So I'd like to get into some of the technicalities soon. But I would like you to explain the difference between a Turing machine, a universal Turing machine,"
},
{
"end_time": 713.797,
"index": 27,
"start_time": 694.104,
"text": " And then a quantum Turing machine and a constructor and a universal constructor. So please. Yeah. So this is a great question because it goes, let's say at the heart of the matter. So a Turing machine is a"
},
{
"end_time": 743.643,
"index": 28,
"start_time": 714.48,
"text": " programmable um machine that so it's a it's an entity which has um you know you can program to do a number of tasks or transformations and these transformations are computations so uh they are a particular kind of transformations that involve um if you like information variables and the classical Turing machine is a Turing machine that operates according to the laws of physics that"
},
{
"end_time": 770.845,
"index": 29,
"start_time": 743.951,
"text": " If you like newton discovered albeit a discretized version of those laws but let's say you know it runs on that kind of physics therefore does not have all the new and very interesting effects that quantum theory led us to discover about a century ago when it was proposed"
},
{
"end_time": 798.097,
"index": 30,
"start_time": 771.715,
"text": " A quantum Turing machine is a programmable computer is a program machine can perform computations that obeys the laws of quantum theory. So instead of having the laws of Newton or discretized version of those, we have quantum theory, which is by the way, the one of the best available explanations of the universe. And I think David told me and I like those, which was quite fun at some point where"
},
{
"end_time": 827.039,
"index": 31,
"start_time": 799.019,
"text": " you know, telling me about how he thought about this universal quantum computer idea initially was then was that he was discussing with someone. And I think it was it occurred to him that somehow the laws of well, the laws of a standard Turing machine, a classical Turing machine were running on the wrong physics in a sense, because they were using an obsolete type of physics, right? Newton's laws, if you like."
},
{
"end_time": 835.725,
"index": 32,
"start_time": 827.398,
"text": " Where is the computer scientist should have looked into something that run on the actual."
},
{
"end_time": 860.708,
"index": 33,
"start_time": 836.169,
"text": " I love the physics that are updated now and they obey the sort of scheme of quantum theory. Somehow the idea was to update, upgrade the idea of a classical Turing machine with the right laws of physics, with the laws of physics that we know superseded Newton's laws at the beginning of the last century. Now a universal"
},
{
"end_time": 887.807,
"index": 34,
"start_time": 861.254,
"text": " Turing machine is a computer, so programmable machine that can perform computations that's capable of performing all physically allowed computations. So not just one computational tool or whatever, but if you consider the set of all physically possible computations under a given law of physics, the universal Turing machine in that specific physics"
},
{
"end_time": 909.104,
"index": 35,
"start_time": 888.046,
"text": " is capable of performing all of them. So for example, you can think of a computation being, I don't know, an addition, you know, you can think of addition and multiplication. These are two possible computations. You can imagine two specific entities, one that can add things and another computer that can only multiply."
},
{
"end_time": 935.845,
"index": 36,
"start_time": 909.991,
"text": " Now a universal, like a more universal machine than either of those is one that is a computer that can be programmed to perform either multiplication or addition. And now if you consider all the possible computations, a universal computer is one that can be programmed to perform all of these computations put together. So it's like a multifunctional entity."
},
{
"end_time": 961.596,
"index": 37,
"start_time": 936.203,
"text": " And so provided that you give it the right program, it will perform the right computation. Sorry to interrupt with the computer that someone is listening to this on their iPhone or their desktop. Is that a universal computer? Yes. So that's a universal. It's an approximation of a universal classical Turing machine. Yes. So I think that's what Turing gave us with his ideas was basically the"
},
{
"end_time": 986.015,
"index": 38,
"start_time": 961.852,
"text": " Model the theory, let's say that that powered all of the information technology that we currently use. And the idea is that the quantum universal Turing machine will upgrade these machines when when they you know when when the universal quantum computer comes about in ways that they can perform new algorithms that are based on quantum laws rather than the classical laws of physics."
},
{
"end_time": 1016.493,
"index": 39,
"start_time": 987.022,
"text": " And our constructors are simply, so if you are happy with this idea of programming something to perform a computation, which is what Turing machines are about, constructors bring this concept a level up in the sense that instead of just having computations as the transformations that you're considering in the repertoire of your machine, you have any physical transformation that is conceivable. So,"
},
{
"end_time": 1046.254,
"index": 40,
"start_time": 1017.295,
"text": " in the case of a constructor is a programmable machine that can perform a given task. The task can be a computation. So computers, Turing machines are special cases of constructors, but constructors can be more general. And typically examples of constructors are things like catalysts, computers, as I said, heat engines, 3D printers,"
},
{
"end_time": 1076.527,
"index": 41,
"start_time": 1047.022,
"text": " Also some machines that can be programmed to perform a transformation, a physical transformation, and also have the ability that they can perform it and stay unchanged in the capacity of performing the transformation again. This is very important. This is true for computers too. There are special cases for constructors because, you know, they perform a computation and then you want them to be able to do it over and over again. And I think the ideal Turing machine should be"
},
{
"end_time": 1106.152,
"index": 42,
"start_time": 1076.8,
"text": " able to do this indefinitely. Likewise, constructors have this feature of being able to perform a task and then repeat this over and over again if given the right input. And so again, programmable constructors are those that can be programmed to perform given tasks. And then a universal constructor is a constructor that has all of the possible tasks in its repertoire. So you can program it to perform any task that is physically allowed."
},
{
"end_time": 1132.927,
"index": 43,
"start_time": 1106.34,
"text": " And there can be a quantum programmable constructor and the quantum universal constructor and possibly a universal constructor that runs on better laws of physics than those we currently know. So maybe post quantum constructors. And so the idea is always the same. Different laws of physics give you different sets of tasks that can be performed just like different laws of physics give you different computations being performable by a Turing machine."
},
{
"end_time": 1162.449,
"index": 44,
"start_time": 1133.353,
"text": " And so, von Neumann's idea was really to extend the scheme of Turing's to other tasks that are not just computations. So thermodynamics transformations are an example, chemical reactions are another example. And von Neumann specifically was concerned about emulating life. And so he noticed that the reason why he thought of this constructor idea was that he noticed that in the Turing machines model,"
},
{
"end_time": 1191.22,
"index": 45,
"start_time": 1163.183,
"text": " There was a gap in the sense that you are not. It's impossible to program a Turing machine, a universal Turing machine, whether it's quantum classical or whatever, to self replicate. So you can program a computer to simulate a self reproducing cell. But if you wanted to program your own computer to create a replica of itself, which would be very convenient,"
},
{
"end_time": 1221.271,
"index": 46,
"start_time": 1191.578,
"text": " Okay, great. So how does one make a difference between what's extremely unlikely and what's impossible?"
},
{
"end_time": 1250.196,
"index": 47,
"start_time": 1221.732,
"text": " So it's my understanding from what you've just said and from the papers I've read is that the traditional way of doing physics is that you have some initial data set and you evolve it forward. That's like the dynamics. And the constructor way of looking at it is, okay, well, actually, let me go back. What you can do is then look at these laws as some causes that produce some effect. And one of those effects may be that entropy tends to increase."
},
{
"end_time": 1265.691,
"index": 48,
"start_time": 1250.913,
"text": " Okay, so in other words, you can derive thermodynamics from statistical mechanics. So it seems like constructor theory is working backward by looking at the effects and then stating those as laws rather than deriving them."
},
{
"end_time": 1288.37,
"index": 49,
"start_time": 1266.271,
"text": " So when I hear you say, look, entropy doesn't, or the heat engine entropy doesn't increase, but well, entropy is not likely to increase. So at what point do we make a cutoff between the likeliness and saying that something's impossible? This Marshawn beast mode Lynch prize pick is making sports season even more fun on projects, whether you're a football fan, a basketball fan, it always feel good to be right."
},
{
"end_time": 1315.998,
"index": 50,
"start_time": 1288.609,
"text": " Right now, new users get $50 instantly in lineups when you play your first $5. The app is simple to use. Pick two or more players. Pick more or less on their stat projections. Anything from touchdown to threes. And if you write, you can win big. Mix and match players from any sport on PrizePix, America's number one daily fantasy sports app. PrizePix is available in 40 plus states, including California, Texas,"
},
{
"end_time": 1343.148,
"index": 51,
"start_time": 1316.271,
"text": " This is a great question."
},
{
"end_time": 1372.927,
"index": 52,
"start_time": 1344.07,
"text": " The so let's first consider the fact that when you have to, when you say that something is possible or impossible, you're not directly referring to likelihood or probabilities for it to happen. So somehow probabilities are not there in the foundations of constructed theory. And they come as say derivative statements, approximate things, but somehow they're not necessarily built. In fact, they're not built at all in the foundations of the theory. And this is a plus in a sense. So when you say that something is"
},
{
"end_time": 1398.712,
"index": 53,
"start_time": 1373.49,
"text": " a task is impossible. What you mean is that there is a law that forbids the fact that this transformation, this physical transformation that's referred to by the task is brought about to arbitrarily high accuracy by a constructor. So by this device that"
},
{
"end_time": 1428.319,
"index": 54,
"start_time": 1399.224,
"text": " can operate in a cycle by returning the substrate that you gave to the constructor in the right input state in the correct output state. And so if the laws of physics say there is some fundamental limit beyond which we can go as far as this transformation is concerned, meaning there cannot be a full cycle that can operate this transformation and the return itself to the original"
},
{
"end_time": 1452.91,
"index": 55,
"start_time": 1428.575,
"text": " state of affairs, just like a catalyst would do. That's the case when the task is impossible. A very simple example is the task of changing the energy of a substrate. So if you want to change the energy of a battery, for example, so you know, from low to high, you need so you can't do it with you can't do it with a constructor"
},
{
"end_time": 1471.527,
"index": 56,
"start_time": 1453.507,
"text": " Without any other side effects, because the constructor will have to give up some energy because of conservation of energy would give up some energy to the battery. Yes, yes. Thereby not being able to return itself to the original state of being able to perform the task again to the same degree of accuracy."
},
{
"end_time": 1498.046,
"index": 57,
"start_time": 1472.039,
"text": " So that's an example of a transformation that's actually possible in the sense it can happen. It can occur on the laws of physics because obviously we can use another source of energy to replenish the battery once it's on low. However, this other source will have to be depleted itself, so it cannot be a constructor. So now if you're looking for a constructor that can"
},
{
"end_time": 1523.729,
"index": 58,
"start_time": 1498.49,
"text": " Give you energy without depleting itself, you will have to go to a physics where conservation of energy isn't true. And so it's not the kind of physics that we believe kind of describes our universe. Likewise, you can think of another example of an impossible transformation. That's the case of in quantum theory. We know that we cannot reliably copy"
},
{
"end_time": 1549.462,
"index": 59,
"start_time": 1524.326,
"text": " any two states of a physical system. This is related to the Isenberg's uncertainty principle. The fact that you cannot measure reliably any two observables of a physical system typically is used, you know, we say the velocity and the position of an electron cannot be both measured simultaneously with the same device. Indirectly, this is constructed statement because it says we cannot have a constructor that can copy or"
},
{
"end_time": 1567.193,
"index": 60,
"start_time": 1550.23,
"text": " or measure accurately these two variables without changing itself in some way. And so it's no longer a constructor, right? So these are examples of impossible tasks and not that I haven't talked about probabilities there yet."
},
{
"end_time": 1597.637,
"index": 61,
"start_time": 1568.012,
"text": " And now if you have lots of physics that, let's say, tell you a number of things that are impossible, but the rest, they don't constrain then whatever is left is a possible task. So if you, you know, if you think of a way of expressing a theory in constructors, the ethic terms, you will have a number of statements about tasks being impossible. And those that aren't impossible are possible in the sense that then somehow it's, it's, it's allowed to, to bring about a constructor that can"
},
{
"end_time": 1626.834,
"index": 62,
"start_time": 1598.08,
"text": " perform these tasks. And there is where you can think of somehow approximate constructors. So when you're thinking of a possible task being realized, performed, let's imagine a possible task is, for instance, to copy the two values of a bit. So, you know, zero and one can be copied. We know it is possible. We do it all the time, approximately in computers."
},
{
"end_time": 1654.138,
"index": 63,
"start_time": 1627.295,
"text": " Um, so, you know, you have zero and one and et cetera. Now, um, the fact that the task is possible is, um, a statement about an idealized scenario where you're thinking, okay, that means that, um, there's a number, there are a number of ways of, um, approximating arbitrarily. Well, this behavior of an ideal constructor can copy zero and one when given them. Yes."
},
{
"end_time": 1681.749,
"index": 64,
"start_time": 1654.804,
"text": " Now, of course, if we look at each particular realization of a copier in any of the computers that we have, for example, they will be approximate. So they won't work perfectly in the sense that at some point they may break down, they may incur in errors, et cetera. But this is simply a feature of the fact that our, you know, we're using limited resources to implement each particular realization of a copier."
},
{
"end_time": 1704.531,
"index": 65,
"start_time": 1682.346,
"text": " However, the laws of physics, as far as we know, don't put any limit on how well we can copy. So it means that for each of these imperfect constructors that are approximate, we can work a bit harder, you know, put a bit more resources into the into the particular device that we have, make it better so we can meet a better accuracy target if we want to."
},
{
"end_time": 1732.449,
"index": 66,
"start_time": 1705.162,
"text": " And so the fact that the task is possible simply means there isn't a limitation beyond which we cannot go as far as accuracy is concerned for this task, as far as we know. So these are very different statements from the statement that something is unlikely. So a transformation can be unlikely or more likely depending on the kind of statement you're looking at. But that would mean simply, for example, that in the standard way of thinking about physics,"
},
{
"end_time": 1762.978,
"index": 67,
"start_time": 1733.012,
"text": " there is"
},
{
"end_time": 1787.517,
"index": 68,
"start_time": 1763.217,
"text": " If we run the system and those are the initial conditions, if we run repeated experiments, we will see most of the times that trajectory to occurring. But it doesn't mean that necessarily there is a constructor. So this device that can work in a cycle to bring this transformation about. So the fact that the trajectory is very likely doesn't necessarily mean the task associated with it is possible."
},
{
"end_time": 1815.23,
"index": 69,
"start_time": 1787.961,
"text": " Likewise, the fact that the transformation is unlikely doesn't necessarily mean that the task is concerned. So for example, you could say given the initial conditions we have, some transformations that are currently occurring really very reliably in some laboratories, I don't know, in CERN or something, they're very unlikely compared to the standard"
},
{
"end_time": 1840.896,
"index": 70,
"start_time": 1815.828,
"text": " physical processes occur naturally in the universe. However, they are possible tasks because we can simply harness enough energy and enough sort of devices that compose the CERN labs, if you like, and we can actually reliably bring those transformations about. So even though some trajectories are unlikely, they can correspond to possible tasks."
},
{
"end_time": 1864.155,
"index": 71,
"start_time": 1841.561,
"text": " In fact, most of the things we do in laboratories, even in quantum computing laboratories, are very unlikely trajectories for certain entities, certain charges or whatever field, etc. And yet we can bring them about really accurately simply because we are following some sort of program to implement these things in the laboratory."
},
{
"end_time": 1894.189,
"index": 72,
"start_time": 1864.957,
"text": " So I guess this is a difference between something being likely and something being possible and likewise unlikely and impossible. I see. OK, let me see if I have the terminology correct. There's something called resource theories and I understand that constructors theory is not a resource theory, but resource theories also deal with tasks and then ingredients that you put together to make some input transformation, some output. Let's say we have this cup and we have a bucket of water, a full bucket of water, and I want to fill this cup so then we can combine"
},
{
"end_time": 1919.804,
"index": 73,
"start_time": 1894.428,
"text": " them to make the task full cup of water after the bucket has poured in some water into it yeah but it's my understanding that this bucket wouldn't be a constructor because this bucket runs out of water it's not something you can keep repeating yes yeah okay so then is this realistic then for a constructor to exist there's no infinite bucket of water that exists for instance is that realistic how do you think about this"
},
{
"end_time": 1949.667,
"index": 74,
"start_time": 1921.34,
"text": " Well, that simply means the transformations of changing the content of a battery or whatever. If you're talking about a concert quantity, I think in your example, if you like, you can substitute your example with the idea of energy and the battery, because then the conservation of energy is the thing that puts a limit on what you can do. So in the case of the energy conservation and changing the"
},
{
"end_time": 1979.121,
"index": 75,
"start_time": 1950.162,
"text": " energy value of a battery. There is a different task that is possible, which is that it's the is the task of transferring some energy from one subsystem to another. So that is a possible task. Meaning there can be a constructed, you know, I think that reliably, you know, if you have a if you have an iPhone or whatever smartphone that's run out of juice,"
},
{
"end_time": 2005.794,
"index": 76,
"start_time": 1979.497,
"text": " You can plug it in as power supply. And if you consider the joint system on the power supply and the smartphone, on that system, there is a possible task that can be performed and simply means you're transferring some energy from one side to the other. So it's like, if you want to think about it, not in terms of batteries and charges, you can think of it in terms of a seesaw."
},
{
"end_time": 2035.265,
"index": 77,
"start_time": 2006.647,
"text": " So you have like a seesaw with two weights and they can move like this. And the task of changing the relative positions of these two entities is possible. However, changing the position of one side is not a possible task by itself because you would have to use energy to do that. So you can still talk about the fact that a charger is possible or a seesaw is possible."
},
{
"end_time": 2061.135,
"index": 78,
"start_time": 2035.759,
"text": " By considering the joint system of the thing that you want to recharge and charge the battery supply or if you like in the case of the seesaw of both sides of the seesaw. So it's completely I would say it's completely fine to talk about it in those terms. And in a way it's more insightful because it tells you it tells you somehow"
},
{
"end_time": 2081.391,
"index": 79,
"start_time": 2061.459,
"text": " Where is it that the constraints are right so the reason why we need a power supply is because the you know the fact that we care about the fact that battery runs out of juice is simply because there isn't a. What we need to supply the energy from somewhere else once it's gone."
},
{
"end_time": 2095.094,
"index": 80,
"start_time": 2082.005,
"text": " And the reason why we need to do that is the conservation of energy. After all, the fact that overall in the universe, the energy is conserved. And so whenever you change energy in one subsystem, you have to change it somewhere else as well."
},
{
"end_time": 2124.872,
"index": 81,
"start_time": 2095.811,
"text": " And the interesting thing is that you can explain then some limitations of what you can do, for example, with a heat engine or like, you know, any kind of engine that runs on the laws of thermodynamics as we know them in terms of the fact that energy is conserved. So that's the explanatory power of the law of conservation of energy, which you can express in constructive terms by saying the energy, changing the energy of the substrate is impossible."
},
{
"end_time": 2152.688,
"index": 82,
"start_time": 2125.742,
"text": " Um, and so that's the content of the theory. And in a sense, I think unlike what you mentioned resource theory, the, the, the difference between say was constructed theory doesn't and resource theory is simply that, um, well, there's some technical difference, but I think the most important difference is that resource theory is more like a framework where you can express existing, uh, dynamical laws or cement with some symmetries."
},
{
"end_time": 2181.34,
"index": 83,
"start_time": 2153.285,
"text": " in this language of transformations being allowed or disallowed. And then sometimes they also care about the fact that the transformations are performed reliably. And then they talk about the catalyst, which has some overlap with constructors. But the main difference is that they don't have principles of their own. So resource theory is not a physical theory. It's a, it's a, it's a framework to express physical theories. Whereas constructors theory has principle principles of its own. So it's, it's, it's, it's an attempt to have"
},
{
"end_time": 2211.988,
"index": 84,
"start_time": 2182.5,
"text": " Actual laws of physics, in addition to those we have currently that can supplement the dynamical laws and tell us more about the universe. So the laws of information are an example. The new laws of thermodynamics that we formulated are another example. So somehow it goes beyond the physics we know currently in the hope of having new predictions, ultimately explanations as well, but also ultimately having informing some new tests that can be somehow testing laboratory sort of changed."
},
{
"end_time": 2241.817,
"index": 85,
"start_time": 2212.381,
"text": " You know, it can provide some sort of extra predictions, extra tests compared to what we currently can do with the laws that we know. When you're repeating a task, are you doing so using a constructor or the constructor? So in other words, are you reusing the same constructor or are you pulling from a different constructor every single time? No, so it's the in the in the ideal case is the same. So so let's talk about the ideal case."
},
{
"end_time": 2267.637,
"index": 86,
"start_time": 2242.483,
"text": " It's a bit like a heat engine is the same constructor. So the idea is you got the same constructor like a fridge and you want to call the. I don't know. Kind of coffee of sorts, sure. And so there's a fridge in the power supply. Yeah, exactly. And and it should be so once you do it with one object, you would like to take it out, enjoy and then put a new one inside the same fridge."
},
{
"end_time": 2298.302,
"index": 87,
"start_time": 2268.387,
"text": " Ultimately, a particular fridge, simply because it's imperfect, will at some point break. But the point is that you can build a better one that lasts longer. And as far as we know, there is no limitation to how well you can approximate the perfect fridge, considering also the power supply that comes with it. Whereas for other things like copiers for quantum states, for states that are not orthogonal in quantum theory, so that this perfect measure for, say, position and velocity of an electron"
},
{
"end_time": 2325.794,
"index": 88,
"start_time": 2298.882,
"text": " It's not that any actual instance of this measure is imperfect. It's really that you cannot construct a machine of that kind. So there is a limitation to the accuracy that this measure can work. And this accuracy simply cannot be increased beyond a certain value. And this is a law of physics that's built into quantum theory. So these are the two different statements."
},
{
"end_time": 2354.599,
"index": 89,
"start_time": 2326.374,
"text": " One statement is any particular instance of a real constructor, a real approximate constructor will at some point break down, but we can perform, we can build a better one. And that's the case when the task is possible. When the task is impossible, like in the case of measuring position and velocity of a particle, of a quantum particle, a measure for those two"
},
{
"end_time": 2378.575,
"index": 90,
"start_time": 2355.572,
"text": " simply cannot exist, which means that you can build very poor measures of those two things that will be wrong most of the times when they're trying to measure both position of momentum, position of velocity, and they cannot improve beyond the certain accuracy. There's like a finite limit beyond which they can't go. And these are two very different situations."
},
{
"end_time": 2407.432,
"index": 91,
"start_time": 2380.026,
"text": " Do you think of constructor theory as a law of physics or as a framework for explaining physics? Like, is it more general than just physics? Is it a paradigm in science rather than a theory of science? Is it a way of going about investigating? I'd say it's both in the sense that so it has some laws of its own that are formulated in the way that I said. So stating what transformations are possible and impossible. But also"
},
{
"end_time": 2435.333,
"index": 92,
"start_time": 2408.439,
"text": " because it's somehow phrasing this different way from the standard traditional dynamical law plus initial condition type of approach is also a new paradigm. So it's not just, um, as a physical theory with new laws in it, it also has, uh, the value of being a new framework or a new language to express laws of physics. So it has both components, but the most important one to me is I think, um,"
},
{
"end_time": 2461.954,
"index": 93,
"start_time": 2436.561,
"text": " Well, they're both important, but somehow the one that that intrigues me the most is the fact that it should ultimately allow us to say more than if it pans out as we expect more than what the current laws say. So it has a physical content of its own, which is non-trivial. Otherwise it would be just a framework to rephrase things that we already know, which could also be interesting, but somehow perhaps is less interesting than say this, um,"
},
{
"end_time": 2489.07,
"index": 94,
"start_time": 2462.619,
"text": " You hope to have as an emergent theory, quantum theory and general relativity. Yes. So yeah, so I, so another aspect of constructed theory is that perhaps this is a compliment supplement to the previous question you asked. So the,"
},
{
"end_time": 2518.626,
"index": 95,
"start_time": 2491.493,
"text": " The way in which constructive theory works is that it doesn't pin down with its principles one specific dynamical theory. So we are hoping that, and I think we demonstrate this for quantum theory at least, and we're in the process of doing this for general relativity, that both quantum theory and general relativity are compatible with constructive theory's principles. So for instance, if you take the principles about information, we know they are compatible with quantum theory."
},
{
"end_time": 2547.534,
"index": 96,
"start_time": 2519.428,
"text": " and we also have arguments to say that they are compatible with general relativity. And in that sense, they are nice because even if you are maybe skeptical about the fact that these principles are really fundamental laws, so you can be agnostic about whether this is a better way to formulate the laws of physics, you can still find them useful because you can still find these principles of constructor theory very useful because if they are things that are obeyed by both"
},
{
"end_time": 2576.049,
"index": 97,
"start_time": 2548.319,
"text": " Quantum theory and general activity that we know don't, um, you know, don't go together as, as theories themselves, because general activity is a classical theory. It doesn't have quantum effects, whereas quantum theory is quantum. Uh, you can appeal to these more general principles of constructed theories to provide explanations and make predictions in a context where both general activity and quantum theory apply, but we don't know how to put the two theories together."
},
{
"end_time": 2602.705,
"index": 98,
"start_time": 2576.493,
"text": " But then we can appeal to these more general principles that do apply in that regime. And that's very much of interest for testing quantum gravity, because that's exactly the regime where we know that maybe one of the quantum theories of gravity that have been proposed may apply, but we don't know which one is the right one. And so having these principles are more general is very useful because you can appeal to them. And the fact that they work both for GR, for general relativity and quantum theory is a plus."
},
{
"end_time": 2614.155,
"index": 99,
"start_time": 2603.063,
"text": " I'm thinking of quantum theory also as a bridge. These principles could be useful to guess a theory that goes beyond quantum theory and GR."
},
{
"end_time": 2643.046,
"index": 100,
"start_time": 2614.394,
"text": " as well as to help us find predictions or experiments that can test the realm where this theory is relevant. So this realm with, for example, testing quantum gravity effects is, well, it's one of the applications of these principles of constructive theory. And that's the stuff I was mentioning earlier I've done with Vlatko. So thinking of constructor theory as some high energy theory that in the low energy limit reproduces the standard model or GR"
},
{
"end_time": 2644.923,
"index": 101,
"start_time": 2643.524,
"text": " is the wrong way of thinking about it."
},
{
"end_time": 2675.401,
"index": 102,
"start_time": 2647.142,
"text": " Yes, I think that's the standard way of going about some theory of everything. That's actually a great thing you said there, because that's the standard way to go about this thing of finding quantum gravity theories, right? So you think, well, you know, I've got these two things, you know, I've got quantum theory, no relativistic quantum theory, and then I have relativistic theories like either GR, general relativity or special relativity."
},
{
"end_time": 2699.974,
"index": 103,
"start_time": 2675.964,
"text": " Then let's find a way to put them together mathematically, right? And then there are proposals how to do that. We have quantum field theory. Then we have some quantum theories of gravity that can work in a low energy regime. Then there are those that actually work at higher energy, et cetera. And each of those will give you a prediction."
},
{
"end_time": 2724.138,
"index": 104,
"start_time": 2700.674,
"text": " Unfortunately, most of the predictions can be tested for the quantum gravity predictions are very difficult to test. And then so what you do is that you say, Okay, well, let's just look at some regimes that are experimentally accessible. So that's the kind of logic that you have when you present those theories. And you're hoping that one of them will be, you will find an experiment that corroborates one of those quantum theories of gravity and refutes"
},
{
"end_time": 2753.285,
"index": 105,
"start_time": 2724.906,
"text": " effectively the classical theory of gravity, which is general relativity. So you would like ultimately some sign of the fact that gravity is quantum. So it doesn't obey general relativity after all, because general relativity is a classical theory. Now, in constructive theories, we are taking a different approach. We are considering the set of, if you like, symmetries or constraints that"
},
{
"end_time": 2783.114,
"index": 106,
"start_time": 2754.258,
"text": " both quantum theory and general activity satisfied together. So it's an exercise of saying, okay, I don't really look at the specific dynamical laws that these two theories have. But I'm trying to extract some deeper symmetry that they both agree on. For instance, the fact that both allow for things like observables. There is a concept of an observable both in quantum theory as well as in general activity."
},
{
"end_time": 2807.363,
"index": 107,
"start_time": 2783.865,
"text": " And these observables obey certain laws of information theory and constructively we can express those laws. And then by looking at this common area of, let's say, agreement of the two theories, you can consider this deeper structure where the two theories agree, which requires you to"
},
{
"end_time": 2833.729,
"index": 108,
"start_time": 2807.927,
"text": " forget about most of the formal details of the two theories. So you will be throwing away various aspects of both quantum theory and general relativity. There are specific formalisms, but you're looking at this deeper structure where they do agree. And for example, the information theoretic structure that concerns observables is something that the two theories agree on. So the fact that there are local observables that you can measure some of these observables and so on is something that the two theories agree on."
},
{
"end_time": 2863.456,
"index": 109,
"start_time": 2834.821,
"text": " Then, of course, general relativity is classical, quantum theory isn't, but I think there is a fundamental structure of observables that they share. They also share things like locality and other features of information theory that are in common. And you use these things, these common constraints, to make some prediction about these regimes where"
},
{
"end_time": 2892.125,
"index": 110,
"start_time": 2864.053,
"text": " quantum system is in interaction with gravity and this actually is very powerful because it allows you to imagine experiments that are in the low energy regime but they allow you to extract quantum features of gravity in a way that's easier than ways that somehow were proposed before to test specific quantum theory of gravity and this is very exciting because it tells us that"
},
{
"end_time": 2919.974,
"index": 111,
"start_time": 2893.08,
"text": " Quantum effects in gravity can actually be easier to capture than it was previously thought. And it's nice that these tests that I'm discussing as part of my work with LAPCA and other collaborators, these tests are really somehow probing a regime where we're not going to very high energies, so it's easier to actually access those regimes."
},
{
"end_time": 2947.739,
"index": 112,
"start_time": 2920.435,
"text": " And they rest on these general principles of information theory. They're very robust and they obey both by general relativity and by quantum theory. So that's maybe the way in which it's nice to think about constructive theory as being relevant to this problem that you mentioned. And you mentioned earlier post quantum, the word post quantum, which rings a bell to me with Jonathan Oppenheim's stochastic gravity. And I believe you're both in the same university, so maybe the same departments. Is that the case?"
},
{
"end_time": 2975.794,
"index": 113,
"start_time": 2948.2,
"text": " I think he's at UCL. We are in the same kind of, but we are interested in the same topic to some degree. I guess quantum foundations, broadly speaking. I think he's in UCL and I'm in Oxford, but it doesn't matter. It's no, no, no, it's, it's, it's true that, that we have some common interests. Which is British accent. Yeah."
},
{
"end_time": 3006.664,
"index": 114,
"start_time": 2977.125,
"text": " And yeah, I think, yeah, I think approximating a British accents is very difficult for me, but I find my best sort of, you know, blending. But but yeah, I think the, the thing that you mentioned is very relevant because, okay, that's an example of a theory that is specifically going to describe gravity and quantum systems in a way that gravity is classical."
},
{
"end_time": 3032.705,
"index": 115,
"start_time": 3007.278,
"text": " So that's so, you know, broadly speaking, we can divide physicists into two camps. One camp says that gravity is actually quantum and we only, you know, we just have to find the right set of conditions to show that it is quantum, uh, with an experiment. And also we have some, uh, ready theoretical proposals for quantum gravity, which exists. Um, some of them are, um,"
},
{
"end_time": 3057.159,
"index": 116,
"start_time": 3033.473,
"text": " been developed in the past decades and they sort of have some predictions etc. But let's say no matter which particular proposal you favor, I think if you are in this camp your heart is saying that gravity is quantum mechanical after all, whereas general relativity says it's classical. And then there is a different camp that says that gravity is actually classical."
},
{
"end_time": 3070.452,
"index": 117,
"start_time": 3057.688,
"text": " And when it couples to quantum system, various things can occur, but ultimately will cause the quantum system to derail and become somewhat less of a quantum system. So it will become more classical than it should be."
},
{
"end_time": 3099.684,
"index": 118,
"start_time": 3071.049,
"text": " In this camp, you will find not just Jonathan Oppenheim, but many people that have been proposing for four years these dynamical collapse theories. Gerardi, Rimini, Weber, there are all sorts of big names there. Roger Penrose with his collapsed wave function due to gravity. So there are all sorts of great minds have been sort of powering this camp in many different ways."
},
{
"end_time": 3124.65,
"index": 119,
"start_time": 3100.06,
"text": " Which one do you belong to? I belong to the former, so I think gravity is not just, I think gravity is quantum mechanical and I think we just have to wait for the right experiment to be performed. But the exciting thing is that this theory that you mentioned that Jonathan Oppenheim has put forward, together with many other theories of classical gravity that have been proposed over the years,"
},
{
"end_time": 3155.23,
"index": 120,
"start_time": 3126.049,
"text": " can be refuted by this experiment that I was mentioning earlier. And this is what makes this experiment very exciting because it can allow us to find a spot where we can at least tell whether gravity has some quantum features or not. So the test would entail two quantum objects, two masses being"
},
{
"end_time": 3183.626,
"index": 121,
"start_time": 3156.067,
"text": " quantum correlated entangled through gravity. So if gravity is capable of creating these correlations between two masses, then we can use this argument from constructed theory to conclude that this actually the gravity has to have some quantum features. And this is nice because it will allow us to rule out immediately all classical theories of gravity, not just general relativity, but also other proposals like"
},
{
"end_time": 3209.497,
"index": 122,
"start_time": 3184.019,
"text": " those I mentioned earlier, the collapse theories, quantum field theory in curved space time, which is another theory where gravity is classical, Jonathan Oppenheim's classical gravity theory and many others. And I think this is very exciting because it's something that hasn't been done so far. And that's the reason why there are still these proposals to say the gravity is classical after all. And it's quite, if you think about it, is really a very important issue at the heart of physics because"
},
{
"end_time": 3240.503,
"index": 123,
"start_time": 3210.794,
"text": " In order to find a good quantum theory of gravity, you need to be motivated that gravity has to be quantum. But if, say, part of the physics community is already even doubting that gravity is quantum, then there isn't much of an incentive to look into quantum theory of gravity. So somehow it will really be important to do this experiment because it will allow us to at least say, OK, now we can give up on this idea that gravity is classical. Let's really get on with it and try to find the right quantum theory of gravity. We don't even have that confirmation."
},
{
"end_time": 3259.94,
"index": 124,
"start_time": 3241.084,
"text": " Experimentally speaking, so it's really nice to have this kind of experiment out there and it's something that people are working on to to actually implement these days So it's really it's really an exciting thing How do we know when to take a limiting theorem like a no-go theorem at one level? So for instance, you said the no cloning theorem before and then apply that"
},
{
"end_time": 3281.647,
"index": 125,
"start_time": 3260.128,
"text": " under constructor theory when we're already thinking that it may not be the case that the laws of quantum mechanics or quantum field theory are the final laws there may be something else that's underneath it. Why are we taking what's a no-go theorem up here and applying it to something that's more fundamental?"
},
{
"end_time": 3314.121,
"index": 126,
"start_time": 3285.333,
"text": " That is a great question and I think it's part of this"
},
{
"end_time": 3339.428,
"index": 127,
"start_time": 3314.565,
"text": " Search for the more fundamental, the deeper structure of quantum theory. So. So if you take quantum theories, it is there are lots of features in it and they are all packaged into the same into the formalism. So if you think for those who are not maybe specialist, the mathematics is a very powerful language."
},
{
"end_time": 3358.848,
"index": 128,
"start_time": 3340.009,
"text": " When you write an equation of motion like in quantum theory, it gives you lots of things just condensed in this equation without giving much depth when you're looking at all of these features. Some of them are deeper than others."
},
{
"end_time": 3385.503,
"index": 129,
"start_time": 3359.36,
"text": " For example, you can have particular features of the mathematical formalism that you're using to express the dynamical law of quantum theories, which may be parochial in the sense that they happen to be relevant for quantum theory, but they're not really fundamental. And then there are some other features to do with the fact that the laws are local,"
},
{
"end_time": 3412.824,
"index": 130,
"start_time": 3386.271,
"text": " That they are, for example, one to one. So they they map, you know, a set of states. There are not two states are going to the same state. So, you know, you keep different states into different states. That's not that's an important it's called logical reversibility. That's one thing that comes for free in the laws of quantum theory. But it's a deeper feature of them that also shared by classical"
},
{
"end_time": 3437.568,
"index": 131,
"start_time": 3413.336,
"text": " And then there are things like what you just mentioned now, the no cloning theorem, which is a thing that you can prove mathematically from the laws of quantum theory, but have a deeper essence in the sense they are part of the of the set of constraints that power the information"
},
{
"end_time": 3468.422,
"index": 132,
"start_time": 3438.524,
"text": " Theoretic structure of quantum information or quantum systems. And so by using an information theoretic perspective, you can see that this no cloning theorem is not just a mathematical feature that happens to be true of specifically quantum theory, but it's a thing that holds promise for being a general feature, something that even if quantum theory turns out to be wrong, so you have to modify the formalism and so on."
},
{
"end_time": 3496.493,
"index": 133,
"start_time": 3469.121,
"text": " It's very likely that this feature of not being able to copy the different states that don't belong to the same physical observable is a feature that will stay. So whatever modification you do to the mathematics of quantum theory. It's very likely that it will be or it's inevitable that it will be. Well, it's difficult to say it's inevitable because these things are"
},
{
"end_time": 3518.234,
"index": 134,
"start_time": 3496.92,
"text": " matter if you like you know as a physicist you also have a matter of taste if you like you know you're thinking of things according to your own philosophy and and so on but i would say that there's there are lots lots of good arguments to expect that this feature will stay uh simply because um it corresponds to the fact that some transformations are impossible"
},
{
"end_time": 3529.514,
"index": 135,
"start_time": 3518.575,
"text": " We know that they're impossible. We've even done a number of tests that actually indirectly test this feature. And so it's an"
},
{
"end_time": 3556.715,
"index": 136,
"start_time": 3529.991,
"text": " If you like, it's to do with the operational information theoretic structure of quantum theory rather than with the specific law of quantum theory. And so it's very natural to imagine that the next theory will conserve these features and maybe have more that are even more exotic and exciting. Just like in the case of classical physics, the fact that"
},
{
"end_time": 3585.538,
"index": 137,
"start_time": 3557.125,
"text": " So if you think of classical physics and quantum theory, the information theoretic structure of classical physics has been maintained within quantum theory, but then there are extra features as we know. So the fact that, you know, you can have an observable in classical physics is also true in quantum theory. It's just that now you have extra properties of these observables that are even more interesting or exotic. And so it's reasonable to expect"
},
{
"end_time": 3611.561,
"index": 138,
"start_time": 3585.981,
"text": " You can make some arguments. I see in fact that these features that have to do with, for example, with no cloning, that is a constraint on what you cannot do with certain information theoretic variables will stay in the next theory. And maybe more constraints will come along and perhaps more interesting properties will be there, but they want undo this change from classical physics."
},
{
"end_time": 3631.288,
"index": 139,
"start_time": 3612.005,
"text": " So to me, it's very unnatural to think that we'll go back to the structure of classical physics where, say, you can clone any state. And so in a sense, this theory, the idea of taking some of these properties of quantum systems,"
},
{
"end_time": 3654.548,
"index": 140,
"start_time": 3631.988,
"text": " their information theoretic versions as general guidelines to describe post quantum systems is a guess, but I think it's a well-informed one. So somehow we are thinking this is how it's going to pan out. But I have to say, many people would disagree. So there's a lot of debate there. It's quite hot as a topic."
},
{
"end_time": 3682.09,
"index": 141,
"start_time": 3654.821,
"text": " And this may be why we can't make progress in certain directions because there are two diametrically opposed ways of looking at things in these physics circles. And that's usually where the fun part of science is. Yeah, that's right. Yeah, exactly. I still want to get to some of your papers, your recent ones. You have a couple on ghost particles and how you can possibly even detect them, which I would love to know more about because it's my understanding that they're undetectable by their nature."
},
{
"end_time": 3710.555,
"index": 142,
"start_time": 3682.5,
"text": " So we're going to get to that, but I believe in one of those papers, you said that there were inequivalent representations of QFT and that that not only has some interpretive issues for philosophers, but also for the mathematicians working on curved spaces or curved space times. So can you please explain what you mean by that? And does this pose a problem to all theories of quantum gravity, even string theory? Right. So, um,"
},
{
"end_time": 3736.169,
"index": 143,
"start_time": 3711.869,
"text": " I don't think that this is necessary. So starting from the last bit of the question, I don't think this is necessarily a problem for quantum gravity theories. So the way this is some work I've done with Vlatko, the way we wrote this paper, this couple of papers, was to investigate some foundations of quantum field theory."
},
{
"end_time": 3764.07,
"index": 144,
"start_time": 3737.312,
"text": " and also of the theory of gravity that's called linear quantum gravity, which is a low energy approximation that on which all of the quantum gravity proposals we now converge. So string theory, loop quantum gravity, et cetera. I agree with this theory that is basically a field theory for gravity, quantum field theory for gravity in the low energy regime."
},
{
"end_time": 3794.138,
"index": 145,
"start_time": 3765.196,
"text": " And all of this was informed by this experiment that I mentioned earlier, and by also some more constructive questions that have to do with what counts as an observable in field theory, both in quantum field theory for electromagnetism, so for the theory of light and for the theory of gravity in this low energy regime. And there were some surprising answers to this question."
},
{
"end_time": 3824.462,
"index": 146,
"start_time": 3794.804,
"text": " And this fits into a more into a broader agenda, if you like, or a broader philosophical take that I have on things. And that I think Vladko also resonates with me on as far as some aspects are concerned, which is that so quantum field theory is a theory that has lots of issues. So it's a problematic theory and in fact was proposed by those who invented it."
},
{
"end_time": 3855.043,
"index": 147,
"start_time": 3825.077,
"text": " more like as a sort of recipe for making calculations as an approximation in, you know, in the while waiting for a better theory. Exactly. Yeah. However, what happened during the years is that while the initial, you know, the founding fathers of quantum field theory did know that this was a sort of collection of tricks to make calculations, but didn't have strong foundations, philosophy and theoretically, the next generation of physicists, which includes"
},
{
"end_time": 3869.923,
"index": 148,
"start_time": 3856.323,
"text": " Many, and finally, also myself, we've somehow forgotten about this fact. And so we are using it to calculate all sorts of things successfully, but somehow we've lost sight of the fact that the foundations are shaky."
},
{
"end_time": 3889.599,
"index": 149,
"start_time": 3870.913,
"text": " So this is an attempt to look into the foundations, go back to look into the foundations. And I think there are many other people who are doing this. So not everyone has this view. But I think the prevailing view is the quantum field theory is all right as it is. And I think that isn't so. And these papers were"
},
{
"end_time": 3920.418,
"index": 150,
"start_time": 3890.589,
"text": " Just as a quick point, when you say the foundations, some people who say I work on the foundations of quantum theory, they mean the foundations of quantum mechanics most of the time. They don't mean I work on the foundations of quantum field theory, which is different. Correct. Yeah, that's right. Yeah, that is true. And so quantum mechanics isn't explicitly relativistic."
},
{
"end_time": 3948.046,
"index": 151,
"start_time": 3920.879,
"text": " So if you want to put it together, at least with special relativity, you need to modify it. And I think quantum field theory is exactly what, you know, is an attempt to do that. And the interesting thing is that even though quantum, no relativistic quantum mechanics is, from the information theory point of view, is more or less equivalent to quantum field theory,"
},
{
"end_time": 3965.725,
"index": 152,
"start_time": 3948.439,
"text": " And also it's also local, doesn't violate, doesn't allow you to signal faster than light and all of these things. It still doesn't have all the features that are satisfactory as far as special relativity at least is concerned. So that's why you have to upgrade the theory to quantum field theory."
},
{
"end_time": 3983.302,
"index": 153,
"start_time": 3966.323,
"text": " However, when you do that, there are lots more problems that come in that occur and these problems are maybe not so important for say making predictions about particle physics and that kind of enterprise which is going well and everyone is happy with what's going on there."
},
{
"end_time": 4006.698,
"index": 154,
"start_time": 3983.302,
"text": " To some degree, but they are still important for the foundations because they inform the way we think of the next theory of quantum gravity itself, et cetera. And so that's that's why we are dissatisfied with it. And we would like to make it, you know, to draw attention to the fact that there is a problem. Those two papers specifically. We're looking at to at so so at both gravity and"
},
{
"end_time": 4036.561,
"index": 155,
"start_time": 4007.312,
"text": " the electromagnetic case, but we focus on the electromagnetic case as an example, both because it's mathematically simpler and also it allows you to make the case in a more transparent way. And to cut a long story short, the way we think about this is that when you try to construct the quantum field theory for the electromagnetic field,"
},
{
"end_time": 4067.022,
"index": 156,
"start_time": 4037.637,
"text": " The usual procedure is that you start from the classical theory of Maxwell's equations, if you like, and then you apply what in jargon is called a quantization procedure. So you can think of it as a sort of machine that you are, you just got the handle, you put in a theory that's classical and out comes something that's quantum. So it's a, it's a procedure that's been put together by various people in a few decades ago. And"
},
{
"end_time": 4097.21,
"index": 157,
"start_time": 4067.841,
"text": " The problem with that is that you can follow different paths to perform this quantization. And even though they all agree on certain observable effects, so experimental outcomes, they are not equivalent physically speaking. So, for example, they have for the electromagnetic field, different ways of describing the quantized electromagnetic field."
},
{
"end_time": 4122.227,
"index": 158,
"start_time": 4097.773,
"text": " have different numbers of subsystems, so that photons are the quantized element of the electromagnetic fields, and different ways of quantizing it leads to different kinds of photons being there. And there is one way of doing this quantization which is explicitly compatible with"
},
{
"end_time": 4148.046,
"index": 159,
"start_time": 4122.534,
"text": " Relativity and this is in jargon is known as choosing a given gauge which is called the Lorentz gauge and when you do this you have basically four kinds of photons that behave quantum mechanically and they are part of your electromagnetic field that is quantized according to this procedure and the typical way of thinking about these things is that"
},
{
"end_time": 4175.486,
"index": 160,
"start_time": 4149.189,
"text": " two of these four kinds of photons are only there as mathematical entities, but they're not really measurable. So they can't really do much. So you can't observe them. They are just tricks, mathematical tricks that are useful to do your calculations, but you shouldn't be able to, not only to detect them directly, so to have a click from these photons,"
},
{
"end_time": 4202.432,
"index": 161,
"start_time": 4176.084,
"text": " but also they shouldn't be detectable otherwise so they're deemed as ghosts because they are there mathematically to help you make calculations but they are not essential in fact there are different ways of quantizing the field the electromagnetic field that only have let's say two kinds of photons not four and so these two ghost modes or two ghost photons are not there yes in those other ways"
},
{
"end_time": 4230.265,
"index": 162,
"start_time": 4203.319,
"text": " And why is it okay? Simply because given that they're not observables, no one cares. And we are all happy that in some other ways of quantizing the field, they're not there. Now, if you so this is, this is the usual story. But the problem is that if you look at the particular kind of experiment that I mentioned earlier, where you have now two charges, not two masses, they interact with"
},
{
"end_time": 4255.674,
"index": 163,
"start_time": 4230.606,
"text": " the standard electrostatic quasi electrostatic sort of Colombian force potential, if you like, and they get entangled through this, through this interaction. So it's a very simple problem, you've got two charges, they're interacting with one another, and they get entangled through this interaction."
},
{
"end_time": 4286.8,
"index": 164,
"start_time": 4257.483,
"text": " If you want a local description of what's going on, meaning a description that satisfies locality at each point of the description, you're forced into using this mode of quantization that uses four kinds of photons. And particularly the ghost photons are very important in the local dynamical description of how the entanglement comes about. So in the papers, in both papers, we make a point that"
},
{
"end_time": 4315.452,
"index": 165,
"start_time": 4287.568,
"text": " There is a way to indirectly detect these ghost photons by looking at the phases that you can create on these charge probes. And this is a thing that hasn't been thought about by people who usually deal with quantum field theory, because they usually think in terms of input-output scattering amplitudes. This is one kind of bit of jargon to just look at"
},
{
"end_time": 4338.285,
"index": 166,
"start_time": 4316.084,
"text": " They look at certain physical processes are very natural to look at in certain contexts, specifically particle and particle physics and quantum field theory in the sort of traditional way. But if you look at quantum theory from the quantum information point of view, where the emphasis is on phases and on things that you can extract out of charges once they interact with the field."
},
{
"end_time": 4368.097,
"index": 167,
"start_time": 4339.309,
"text": " It's very, if you draw, if you use this principle of locality and various other principles to sort of guide your analysis, you will see that measuring features of the two charges among which there is the entanglement between them that's caused by this static interaction is equivalent to measuring accessing these ghost photons as dynamical degrees of freedom. And this is something that"
},
{
"end_time": 4390.674,
"index": 168,
"start_time": 4368.473,
"text": " It's very interesting because somehow it seems to at least contradict the standard way of thinking about these ghost photons. And it can also be the same argument you can carry it out in the gravitational field case in the linear quantum gravity regime. It just is more complicated. So instead of photos, you have gravitons and there are more kinds of gravitons involved."
},
{
"end_time": 4417.892,
"index": 169,
"start_time": 4391.067,
"text": " But the idea is the same. So the idea basically is this, that if you insist on having a local account of what's going on in a very simple quantum information experiment that involves two charges or two masses or even just one charge in the field, you have to somehow come to terms with the fact that these ghost modes are indirectly observable. So you cannot measure a ghost photon in the same way that you can get a click out of a photon. Yes, yes, yes."
},
{
"end_time": 4447.09,
"index": 170,
"start_time": 4418.251,
"text": " But they are important. There are degrees of freedom that can be indirectly uncovered by measuring features of the charges once they interact through the field. It's as if the charges got clouded with the field and then by measuring the charges, you're extracting features of the field. And unfortunately, you're extracting these ghost features and not the ones that are supposed to be measurable. And this is very interesting. I think this is a sort of we are hoping this will cause some disruption in the field also because we have some"
},
{
"end_time": 4473.848,
"index": 171,
"start_time": 4447.449,
"text": " proposal for an experimental test so that the theory that we have can be tested and we're interested to see what happens when we you know when this test can be performed and you can do it both with gravity but also with electromagnetism which is probably easier to do considering you know what we can do experimentally at present man a fantastic name for a theory would be ghost gravity you could write a book on that it's true yeah yeah yeah it's a great it's very"
},
{
"end_time": 4503.319,
"index": 172,
"start_time": 4474.377,
"text": " Yeah, it's very exciting. Yeah, that's true. OK, so what is the physical interpretation then? Because ghosts come about from gauge fixing, which is just something you do to make the math easier akin to if you care about the derivative of a function only, then your regular function can have plus any constant. It'll just go away when you take the derivative and you can set the constant to be whatever you like for whatever reasons, calculation reasons. But the constant goes away when you take the derivative. That's a great point you're making there."
},
{
"end_time": 4531.118,
"index": 173,
"start_time": 4503.882,
"text": " So the gauge fix, so the gauge fixing is exactly what you said. So gauge is corresponds to, if you like these different ways of, okay, let me, let me, let me make a step back and try to explain this a bit more clearly. Okay. Gauges are simply different ways of, of describing the electromagnetic field."
},
{
"end_time": 4558.848,
"index": 174,
"start_time": 4532.244,
"text": " even classically. So even classically, you have different gauges. So, and as you said, it's, they mathematically, they correspond to switching to different variables for your Maxwell for Maxwell's equations. So Maxwell's equations can be written in terms of fields, electromagnetic fields, and they have some classically speaking, but you can also change the variables to express the equations."
},
{
"end_time": 4583.285,
"index": 175,
"start_time": 4559.462,
"text": " And so instead of the fields, you can use these things called potentials, vector and scalar potentials. And the potentials are basically just a different, you know, mathematically speaking, they're just in variables. However, there are many different changes of variables that you can make, and they're all equivalent, they all collapse to the same Maxwell's equations in terms of fields."
},
{
"end_time": 4611.749,
"index": 176,
"start_time": 4584.087,
"text": " And each of these different change of variables that may be useful for computational purposes is called a different gauge. So each of them, each of each gauge has their own have their own names. So there is like Coulomb gauge, Lorentz gauge, scalar gauge. So they have names according to how they were discovered. Right. Now, classically speaking, this is irrelevant, meaning physically relevant in the sense that you can solve the equations with fields or with potentials in one gauge or another, and no one particularly cares about what you're doing."
},
{
"end_time": 4637.449,
"index": 177,
"start_time": 4612.705,
"text": " However, when you quantize the field, things become different because each gauge corresponds to a different quantization procedure, if you like. And so the Lorentz gauge, which is the one that is local and explicitly Lorentz compatible, so it's explicitly compatible with special relativity, leads to these four"
},
{
"end_time": 4656.032,
"index": 178,
"start_time": 4637.875,
"text": " Modes that for ghost sorry for kinds of photos which in jargon are called modes and two of them are ghosts because if you follow different quantizations which start from different gauges not the lorenz gauge but something in some other gauge for example chrome gauge."
},
{
"end_time": 4678.797,
"index": 179,
"start_time": 4656.664,
"text": " There are only two such modes or kinds of photons, and so they are ghosts because they ultimately are somehow usually thought of as being unphysical because they are not present in all gauges. Okay."
},
{
"end_time": 4704.411,
"index": 180,
"start_time": 4679.872,
"text": " So gauge fixing means you pick a particular gauge and that corresponds to some mathematical constraint being there. Usually this constraint is supposed to be irrelevant physically because, as I said, from the classical point of view, all you care about is just Maxwell's equations. And for Maxwell's equations, the potentials may not even exist. And, you know, they're just expressed in terms of fields. Now, the"
},
{
"end_time": 4730.794,
"index": 181,
"start_time": 4705.486,
"text": " The significance of what we discussed, which by the way, I think was discussed by other people. There's a Bernard Kay who's a researcher at York that has also made similar comments recently and in the past other people have also made similar comments, just perhaps motivated by different reasoning. So what we said is that in the quantum case,"
},
{
"end_time": 4756.749,
"index": 182,
"start_time": 4732.312,
"text": " Unlike in the classical case, there are some kinds of experiments that you can perform on the charges, which are the things that you interact with, that you use to somehow extract features out of the field. There are some experiments that you can perform on the charges, if you think that they're quantum, which are not there in the classical case, obviously, because in classical physics, charges are also"
},
{
"end_time": 4785.776,
"index": 183,
"start_time": 4757.654,
"text": " And once you go along with this fact, and if you quantize the field and you want the whole description to be nicely local, et cetera, you are forced to see that the charges get in some situations, they simply get to depend on the degrees of freedom that are supposed to be goals in the Lorentz gauge. And by then measuring the charges in certain situations that are possible now to measure, because we have quantum metrology facilities that allow us to do that,"
},
{
"end_time": 4814.48,
"index": 184,
"start_time": 4786.971,
"text": " You are indirectly accessing these goals. Yes. So in a sense, this idea of gauging variance, which is very important for classical electromagnetism, is also relevant for quantum electromagnetism or quantum electrodynamics. But it doesn't forbid us from, it doesn't impede the realization of these experiments that we discuss in the papers."
},
{
"end_time": 4844.36,
"index": 185,
"start_time": 4815.06,
"text": " And so it forces us to revisit the idea that the ghosts are not measurable. They are not measurable in the sense that they may not be measurable in the standard sense of being measurable. So in usually standard sense means to detect a click. So, you know, you have a photo in submitted and you detect it. This is one way of being measured, of measuring features of the quantum electromagnetic field. But there are other ways of probing it with quantum charges."
},
{
"end_time": 4874.241,
"index": 186,
"start_time": 4845.299,
"text": " Which don't necessarily amount to detecting a photon of the ghost kind, but they amount to some kind of detection. It's not a direct detection of these photons, but they are detecting some quantum features of these ghost modes. So in a way we are suggesting that the idea of measurability and what counts as measurable should be revisited in light of the fact that we can actually perform these experiments. So would an analogy be like the Aronov-Bohm effect where before that the"
},
{
"end_time": 4903.2,
"index": 187,
"start_time": 4874.531,
"text": " Electromagnetic potential was thought to be something that was just mathematical, a convenience. And then afterward, you still don't detect the electromagnetic field directly, but you see its effect on the phases of the electrons that go through a soliton or the outside of a soliton. And that, by the way, was revolutionary. So this paper with you and Vidral, is that correct? Yeah. Let me read its title. Interference in quantum field theory detecting ghosts with phases. And I'll put a link to that in the description."
},
{
"end_time": 4928.524,
"index": 188,
"start_time": 4903.37,
"text": " So this paper, is it suggesting something physical? We think that there's something physical about the electromagnetic field with one of the major pieces of evidence, historically being the Earnhardt-Bohm effect. So with your proposal to detect ghosts, are you saying that something else exists, like something that's choosing a gauge? Is there something else that some other field that is actually in existence that we thought was a mathematical trick?"
},
{
"end_time": 4942.978,
"index": 189,
"start_time": 4930.213,
"text": " Yes, I think this is one way of thinking about what we do. In fact, it's very related to the Arano bomb effect. In the following sense, we've also actually"
},
{
"end_time": 4961.459,
"index": 190,
"start_time": 4943.968,
"text": " thought about the Iron bomb effect and somehow this was part of the way in which we landed on this idea. I see, I see. Because even in the case of the Iron bomb effect, the interaction between the electron and the solenoid is mediated by accident."
},
{
"end_time": 4984.087,
"index": 191,
"start_time": 4961.459,
"text": " They're very similar words, somehow different things, but the solenoid being this thing that generates a magnetic field only somehow ideally along a line and zero field elsewhere."
},
{
"end_time": 5011.954,
"index": 192,
"start_time": 4984.48,
"text": " And in the A-B effect, you have this interesting fact that if you have an electron that goes exactly in the region where the field doesn't appear to be there, it can still be affected by the solenoid because when the solenoid is switched on, the electron has a different phase when it interferes compared to the case when the solenoid is off. So this shows that the electron is actually picking up some signal from the solenoid, even though there is no direct field that's acting on it."
},
{
"end_time": 5040.316,
"index": 193,
"start_time": 5013.49,
"text": " Okay, there the issue was how does how do the solar and the electron communicate? Even though classically speaking, if we describe the whole thing classically, sorry, if you describe the field classically, it looks like there's no field at the point of the electron. And the key to answer that is that even though classically that's true, if you consider the quantum description of the field, you will notice that"
},
{
"end_time": 5060.981,
"index": 194,
"start_time": 5041.237,
"text": " First, there is a back action of the electron on the solenoid. This is something that was already pointed out by Lev Weidman, who anyway used still a classical model for it. And second, more importantly, I think there's the fact that even though the field is zero classically outside of the solenoid,"
},
{
"end_time": 5089.787,
"index": 195,
"start_time": 5061.954,
"text": " a quantum field can never be zero in the strict sense. And so the so it can be zero. Some aspects of the field can be zero, but there is still a quantum feature of the field that's there. And some of the photons that are so there are photons coming in a sense, going back and forth from the solenoid to the electron, vice versa. And these photons happen to be also of the same kind as those that we"
},
{
"end_time": 5110.964,
"index": 196,
"start_time": 5090.435,
"text": " the most"
},
{
"end_time": 5137.892,
"index": 197,
"start_time": 5111.459,
"text": " and this is only clear the fact that everything happens locally and nicely can only be accounted for as far as we know in this in a particular gauge. So somehow this set of results that we mentioned show us on the one hand as you said that some gauges are better than others the Lorentz gauges is"
},
{
"end_time": 5168.336,
"index": 198,
"start_time": 5138.763,
"text": " more accurate in the description of what's going on, because it's explicitly local and also Lorentz covariant. But also there's the fact that in order to have a complete description of what's going on, you have to quantize the field. So you have to have a quantum description of the field. Otherwise, as it happened in the case of Harnov and Bohm, if you stick to a completely classical description of the field and you want the charges and the solenoid"
},
{
"end_time": 5192.585,
"index": 199,
"start_time": 5169.104,
"text": " Sorry, the field is classical but the charge is quantum. You will incur into some issues of the description and Arnold and Bohm somehow were concerned that this effect, as they described it in the semi-classical theory, appeared non-local. And this can be cured by the fact that you quantize the field. So the lesson in this set of experiments is both that"
},
{
"end_time": 5221.305,
"index": 200,
"start_time": 5193.131,
"text": " Some gauges are more accurate than others physically speaking. They're more realistic. They make more sense. So they tell a more coherent story because they're all explicitly local, etc. Lorenz gauge is one of them. The second lesson is that in order to have a local description, you always need to quantize everything in your systems. Otherwise, if you insist on one of the systems being classical and the others not, you will incur some issues typically with locality. And finally, the third lesson is that we will need if we quantize the field and you look at"
},
{
"end_time": 5247.927,
"index": 201,
"start_time": 5221.783,
"text": " some of these effects in specific gauges like the RANS gauge. And you consider all the experiments you can perform on the charges, not just these input, output scattering, and amplitude things that people usually look at in particle physics, for example. You will notice that it's inevitable to have to modify your notion of measurability because even though you cannot detect"
},
{
"end_time": 5276.084,
"index": 202,
"start_time": 5248.592,
"text": " As far as we know, directly these ghost photons, by say having a click in a photo detector, you can indirectly detect their degrees of freedom by making measurements on the charges. An inescapable conclusion of how you analyze the analysis that you can make of the situation in the Lorentz gauge. And so this requires us to just simply enlarge the set of things that we can"
},
{
"end_time": 5306.135,
"index": 203,
"start_time": 5276.783,
"text": " So this, to me, sounds like a departure from constructor theory, but you see it as tied or you see there being implications. Oh, yeah, absolutely. Yeah, I think in my head, all of these things. So if you like, you know, if you look at I'm being very"
},
{
"end_time": 5330.828,
"index": 204,
"start_time": 5306.732,
"text": " Now I'm sort of telling you somehow that things do it the way I think about stuff so I don't know if it's really relevant but the way I think of this is really you know you can look at these papers they look different from each other but I think it's really me trying to understand how to apply this notion of observable that's more general that you have in constructive theory of information so this paper I wrote with David a while ago"
},
{
"end_time": 5358.063,
"index": 205,
"start_time": 5331.22,
"text": " where anything that is copyable, any set of attributes that is copyable can be considered as an observable, whether or not it has some formal features that quantum theory requires, such as being Hermitian operator, et cetera, et cetera. So it's my way of understanding how this concept of an observable can illuminate different parts of quantum field theory or quantum theory more generally."
},
{
"end_time": 5372.995,
"index": 206,
"start_time": 5358.899,
"text": " Especially those cases where we don't have an understanding of what we can really observe. So quantum field theory is an example of a case like that because, as I said, even though it's a quantum theory of fields,"
},
{
"end_time": 5401.084,
"index": 207,
"start_time": 5373.729,
"text": " It has issues with various foundational aspects. And so what counts as an observable there is very confused. And I think the foundations of quantum field theory also are not so clear about what is an observable and what isn't. And so this is a way to, in a way, show the power of, well, I don't know if it shows the power, but let's say it's a way of applying"
},
{
"end_time": 5431.271,
"index": 208,
"start_time": 5401.493,
"text": " These constructed theory notions of observables that are more general than particular notions in quantum theory or quantum field theory to help us find new ways of probing fields compared to what we used to know before. This is a bit technical, but when I was looking up the categorical approach to constructor theory, I saw that what you want to do is you want to divide your space into what's impossible and what's possible and your spaces then or tasks. I believe tasks are the morphisms in a symmetric monoidal category."
},
{
"end_time": 5460.401,
"index": 209,
"start_time": 5431.937,
"text": " and I believe when you divide your space into what's possible, it becomes a subcategory. Do the impossible states also form a subcategory or no? I actually don't know because the category theory can be applied, I'm not an expert at all, but I think it can be applied to a number of things. In fact, it's so flexible, it can be applied to a very broad set of theories in a way, right?"
},
{
"end_time": 5486.476,
"index": 210,
"start_time": 5461.067,
"text": " And I am not sure about the specific answer to this question. I think what is promising there, and I was very happy that some people were interested in category theory, actually decided to do this work, is that it could be that by casting constructed theory in these terms that are, if you like, more modern from the mathematical point of view as well,"
},
{
"end_time": 5511.288,
"index": 211,
"start_time": 5486.971,
"text": " I think there's a chance that some of the theorems we have could be condensed in a very elegant short proof and as well you could also see unifications between things that we think may be distant and not related. I'm really hoping that that continues and that perhaps more"
},
{
"end_time": 5539.019,
"index": 212,
"start_time": 5511.903,
"text": " More people may come along and have more results in that kind of direction. But as far as your question, I'm not so sure. The only thing is that if I sort of understand correctly what you were saying, what allows you to conclude that is the fact that in constructed theory, we have this principle that's called the composition law."
},
{
"end_time": 5568.439,
"index": 213,
"start_time": 5539.633,
"text": " that says when you compose two possible tasks, you still should obtain a possible task. Whereas this isn't true for impossible tasks. Okay, so then it wouldn't form a subcategory. Yeah, they wouldn't. Right. So the idea is that somehow this composition law only holds for possible tasks. And it doesn't for impossible tasks. So in fact, you can have two impossible tasks that compose together give you a possible task."
},
{
"end_time": 5594.531,
"index": 214,
"start_time": 5568.899,
"text": " A typical example is the so imagine again, the task of high of first raising the energy of a system and then lowering by and then the second task is lowering the energy by the same amount. You can compose the two tasks joint, right? So they're both impossible, right? As we said earlier, yes, yes, by themselves."
},
{
"end_time": 5622.295,
"index": 215,
"start_time": 5594.906,
"text": " But if you compose them, you basically have the task of the identity, right? So you stay in the same place. So you go, you know, one task goes up, the other goes down by the same amount is basically the task, the resulting composition gives you the task of staying in the same place, which is a possible task by somehow axiomatic, by assumption. So that's an example of two impossible tasks giving you a possible task. So I'm guessing"
},
{
"end_time": 5651.852,
"index": 216,
"start_time": 5622.756,
"text": " Because there isn't disclosure property, presumably the answer to your question is that the impossible tasks should not have that property. But as I said, I'm not sure. And the principle of locality, that's super important to constructor theory. It's that different than locality in the regular physics sense? No, I think it's so OK. Locality is a very subtle concept and has lots of different meanings in physics."
},
{
"end_time": 5679.462,
"index": 217,
"start_time": 5652.363,
"text": " So in other words, if you have a way to describe your, you know, you have a system that's made of subsystems,"
},
{
"end_time": 5705.93,
"index": 218,
"start_time": 5680.725,
"text": " and you have a transformation that only involves one of these subsystems. It should only be able to change the subsystem in question and not the others. And this applies both to the directly observable things on the other systems and those that are not directly locally observable. And this is a property that is true for quantum theory,"
},
{
"end_time": 5731.869,
"index": 219,
"start_time": 5706.596,
"text": " Non-relativistic quantum theory for quantum field theory is true for general relativity and is true for some forms of, I would say, well, it depends what you're saying, but I think somehow this is true for all the theories that we think are reasonable, even non-relativistic quantum field theory that somehow is not necessarily the correct theory because it's not relativistic."
},
{
"end_time": 5755.674,
"index": 220,
"start_time": 5733.319,
"text": " Now, this is different from the concept of locality that's sometimes referred to as Bell locality, which is a notion that's very important in quantum foundations because it's the property that a theory can be described by a local hidden variable model that uses real value stochastic theories."
},
{
"end_time": 5784.036,
"index": 221,
"start_time": 5756.869,
"text": " And John Bell had this very interesting result, which was a mathematical theorem that says that if you violate with some statistics, certain inequalities, which are called Bell inequalities, you can conclude that your statistics are coming from a theory that is not Bell local, so it cannot be reduced to a local hidden variable model."
},
{
"end_time": 5815.418,
"index": 222,
"start_time": 5785.435,
"text": " And quantum theory is not Bell local, so it's not describable as a local hidden variable theory that uses real value stochastic theories. So in that sense, it's often confused. It's often said that quantum theory is non-local, but what one means there is Bell non-local. So it's the fact that quantum systems can violate Bell inequalities. But the locality we're talking about is really more basic or fundamental. So it's to do with the fact that there is no action at a distance and"
},
{
"end_time": 5845.213,
"index": 223,
"start_time": 5815.998,
"text": " Even quantum systems that are entangled, that violate Bell inequalities, do not allow you to perform this action at a distance. And so that's the feature that constructive theory cares about. And I think this locality is really built in in the foundations of the theory. It's a very important axiom, if you like, or principle. And it's also very important in the context of these experiments that I mentioned earlier, the ones to test quantum gravity and also the"
},
{
"end_time": 5869.872,
"index": 224,
"start_time": 5846.186,
"text": " Electromagnetic discussion we had early with the goals and also the gravity version of those results. So yeah, it's a very important principle. Chiara, what's a no design law, a quote unquote no design law? A no design law is a law that is very important for understanding this is in the origin of life problem."
},
{
"end_time": 5899.275,
"index": 225,
"start_time": 5870.657,
"text": " Frame of, you know, frame of mind. So we're thinking in that camp. Um, so the problem of biology is to one of the problems of evolutionary biology is to explain how, uh, living systems can have come about without, um, that being a designer. So typically I think, you know, Darwin was battling with the idea that somehow there was a"
},
{
"end_time": 5925.077,
"index": 226,
"start_time": 5900.111,
"text": " believe that you had to appeal to a creation sort of, you know, to God being there to design to a designer to create entities are highly complex out of simple beginnings. And so if you don't want to go that way, because you don't want to make that step to believe in such an entity, you're in luck because Darwin and"
},
{
"end_time": 5949.633,
"index": 227,
"start_time": 5925.623,
"text": " All the great biologists who came after him and kind of refined his idea showed us that it's possible to have a complex entity come out to simple beginnings, simple initial conditions that don't include complexity."
},
{
"end_time": 5975.128,
"index": 228,
"start_time": 5950.128,
"text": " simply by waiting for long enough and allowing for these natural selection and basically mutations that are not specific to what you want to get in the end. So you need the mechanism of natural selection to be there and you need the possibility of there being things that can replicate which"
},
{
"end_time": 6002.568,
"index": 229,
"start_time": 5975.828,
"text": " you can call genes in a broad sense, a bit like what you know, with the Dawkins terminology. And so they're not just genetic, they're pieces of, you know, strings of information that can replicate in a stable manner. And then you need the possibility of having errors in these replications that introduce variations that can then be selected by the national selection occurring in a given environment."
},
{
"end_time": 6032.961,
"index": 230,
"start_time": 6004.002,
"text": " Now this, this concept is very, uh, this, this description is usually in biology is given at a high level. So no one really thinks of what about the laws of physics? Um, we, you know, people just think, okay, the laws of physics are in the background. We don't care in biology. That's fine. But somehow when you are trying to imagine a way to reproduce this thing in a laboratory or in a simulator or in a computer or in some sort of dynamical system,"
},
{
"end_time": 6061.152,
"index": 231,
"start_time": 6033.712,
"text": " You have to have a very good understanding of what dynamical interactions you are happy to allow in your simulator to mimic the fact that like in physics, these laws, these dynamical interactions don't have the design of the life that you want to emerge out of your process of evolution."
},
{
"end_time": 6090.282,
"index": 232,
"start_time": 6062.329,
"text": " And this is a very tricky concept. It's tricky even in physics itself and in simulation is even more tricky. But the point of no design laws, the definition of it is really to highlight the importance of the fact that when you're running an argument like Darwin's, you want to make sure that you haven't snuck in some assumptions about the laws that are"
},
{
"end_time": 6116.237,
"index": 233,
"start_time": 6090.367,
"text": " you know, that govern the interaction between microscopic entities in your biological system, which ultimately contain the design of the living system that you create at the end of the natural selection process. So a no design law is a law that isn't specifically designed or crafted to bring about a specific complex entity or a set of specific complex entities. And we know that the laws of physics says"
},
{
"end_time": 6129.548,
"index": 234,
"start_time": 6116.34,
"text": " We believe the laws of physics we know kind of govern the universe. Our current guesses are no design laws in this way because they don't specifically contain symmetries that are especially"
},
{
"end_time": 6151.783,
"index": 235,
"start_time": 6130.06,
"text": " suited for the emergence of a specific form of life. And that's very important to know, because that gives strength to Darwin's argument, if you like. And so if you are concerned with the possibility that there could be a skeptic that might not believe in Darwin's explanation,"
},
{
"end_time": 6181.323,
"index": 236,
"start_time": 6152.159,
"text": " It's good to remind ourselves that the laws of physics that Darwin's presupposes for his own reasoning and neo-Darwinism in general presupposes are no design laws. So nothing is assumed of the dynamical interactions that are used in the sort of Darwin's theory of evolution. And when you then try to, if you're thinking of a way to"
},
{
"end_time": 6211.186,
"index": 237,
"start_time": 6182.432,
"text": " to reproduce this thing in a laboratory. So you think of having or in a computer, it's very important that when you, you know, if you start with simple beginnings and you get complex entities at the end in your simulation, you want to have a criterion and a test or a check away to check that the dynamical interactions that you use in order to show this progression do not already contain the design of what you're getting at the end. So you start with something simple and outcomes, something like an elephant."
},
{
"end_time": 6240.589,
"index": 238,
"start_time": 6212.108,
"text": " Let's say in a simulation in a computer, you want to make sure that the computer doesn't have some rules of interaction between elementary cells that somehow contains the design of this entity that comes at the end. So you don't want to assume what you're trying to prove. That's right. You don't want there to be circularity. I see. Yeah, that's right. So if you were to assume that then you would be circular because it would be like saying, OK, well, I gave you an elephant, but actually I snuck in the program, you know, the design of it. So that's not very surprising."
},
{
"end_time": 6261.305,
"index": 239,
"start_time": 6241.152,
"text": " What's surprising and what's cool in what happened in the biosphere on earth and perhaps elsewhere in the universe we don't know is that as far as we can tell because given that laws are not designed specifically for life, life came about and we have lots of complexity now going around the earth and that's very interesting and somehow"
},
{
"end_time": 6282.363,
"index": 240,
"start_time": 6261.715,
"text": " Okay, great, great, great. Now that we've talked about design law or no design law, what is super information? So super information is a"
},
{
"end_time": 6303.951,
"index": 241,
"start_time": 6283.746,
"text": " Particular kind of, in fact, I should talk about super information media. That's the thing. So it's a generalization of the concept of quantum systems. So it's in the context of the constructed theory of information and the super information media are physical systems that obey the laws of constructed theory."
},
{
"end_time": 6320.316,
"index": 242,
"start_time": 6304.48,
"text": " and have extra properties compared to physical systems like a bit that can only contain classical information. And these extra properties have to do with the fact that not all of the states are copyable, just like quantum systems."
},
{
"end_time": 6346.408,
"index": 243,
"start_time": 6320.896,
"text": " And in our paper with David, we prove that these systems, the super information media have all the qualitative properties of quantum systems. So if you like, you can think of the theories that describe these super information media as a generalization of quantum theory. And a super information medium could be a medium, a physical system that can perform quantum computation."
},
{
"end_time": 6376.459,
"index": 244,
"start_time": 6347.073,
"text": " But may not obey fully quantum theory. And so they are they are like post quantum systems can be described by post quantum theories that obey principles like locality and the interoperability of information. So all of these principles construct the theory. And these super information media are very useful. The theory of these things is very useful in the context that I mentioned earlier, where you have a quantum system interacting with something that like gravity that may or may not be quantum, because a quantum theory of gravity may describe"
},
{
"end_time": 6406.596,
"index": 245,
"start_time": 6376.988,
"text": " I see, I see. Does constructor theory have anything to say about dark energy or dark matter? Okay, this is very speculative. I think we we've remarked a number of times. This is like I said, the speculative level that so these systems are things that may or may not obey quantum theory as we know it. And"
},
{
"end_time": 6431.988,
"index": 246,
"start_time": 6407.261,
"text": " also may require some upgrade of the theories that we currently have in general, even at the kind of mathematical level. And because of that, I think it's very useful to apply to them this theory of super information media and or the constructed theory of information more generally, because they may still obey the principles of information."
},
{
"end_time": 6460.964,
"index": 247,
"start_time": 6432.705,
"text": " and also the principles of information theory that we laid out in our paper put constraints on them. So, for example, the principle of interoperability of information which says that any system that can contain information should also be able to interact with another system that can likewise contain information and you should be able to set up some interactions between them that allow for copying various states. I think this is the thing that can be so this principle"
},
{
"end_time": 6483.183,
"index": 248,
"start_time": 6461.305,
"text": " could be applied to dark matter, for example, and it may turn out that this principle rules out some of the theories about dark matter simply because they have, they somehow violate the principle and they say that some copy-like operations between the dark sector and the non-dark sector are impossible."
},
{
"end_time": 6513.319,
"index": 249,
"start_time": 6483.456,
"text": " So in a sense, I think constructively, you can say a lot about that. And maybe this is something for the future application, but it's definitely something that is on the, you know, something that we may at some point address. Now, given that you're working on what is like a theory of everything, or maybe it's a framework for a toad rather than a toad, but regardless, do you believe a toad to be possible? There are two respects in which I mean that, like, does it exist? So that's one question. And then even if it exists, is it knowable to us?"
},
{
"end_time": 6543.114,
"index": 250,
"start_time": 6516.647,
"text": " So I think, I don't think O'Constructor theory is a theory of everything for the reason that, as I said, it has the ambition to express. So as I said, it has a principle, it has a number of principles that can express some aspects of physics, ideally all of them, but also it may not"
},
{
"end_time": 6571.988,
"index": 251,
"start_time": 6543.507,
"text": " have the features of the dynamical theory. So, so you may not actually directly respond to the standard paradigm of the theory of everything. I think a theory of everything perhaps is not the most fruitful way of thinking about stuff in the sense that it is my philosophical position that we, you know, in any theory, no matter how complete it looks, we"
},
{
"end_time": 6600.879,
"index": 252,
"start_time": 6572.381,
"text": " may be able to find problems. And so these problems will lead to something else. And so the way I think about stuff is more like, I like to think of there being different levels of explanations. Maybe there are infinitely many, and we just keep going from one to another by understanding things more and more. So in that sense, I like this more open-ended way of thinking about science, physics specifically."
},
{
"end_time": 6626.288,
"index": 253,
"start_time": 6601.749,
"text": " Where we just keep digging and sometimes when we dig deeper, we find a different level of explanation that brings some more unification along and we go from one level to another. And I think the reason ultimately, so there isn't really a very strong, I think it's very difficult to convince someone who thinks otherwise that this is the case, but I think I can say that the reason why I like this view is that"
},
{
"end_time": 6656.596,
"index": 254,
"start_time": 6627.21,
"text": " I think it gives you really, as a physicist or as a scientist, it gives you the idea that there'll always be something to work on. And I think that's a fun thing to entertain in your head. And in a sense, the theory of everything somehow suggests this sense of closure that you might achieve at some point. And I think given that we can never know whether what we know is true,"
},
{
"end_time": 6677.056,
"index": 255,
"start_time": 6657.159,
"text": " It seems to me to clash with this epistemological stance that I have that I think it's impossible to actually know whether what you know is really true. All you can hope for is somehow to be able to find problems in what you know and somehow change your view accordingly. So in a way, I think I"
},
{
"end_time": 6703.882,
"index": 256,
"start_time": 6677.602,
"text": " I kind of like this stance on things. I think this is very popular and if you like and I sort of subscribe to that I really like this approach and it's very scientific. I think this how scientists behave more or less even though when they you know, even though sometimes they don't acknowledge that but it seems to me this is kind of very natural way of thinking things. What would be the difference between information and knowledge? So knowledge is a kind of information that"
},
{
"end_time": 6726.22,
"index": 257,
"start_time": 6705.247,
"text": " has extra properties. So information is really some set of states that can be copied to arbitrary high accuracy, whereas knowledge is this extra feature of being information. So a set of states that can be copied, which also are capable of causing transformations to occur on physical systems and to stay"
},
{
"end_time": 6756.203,
"index": 258,
"start_time": 6726.698,
"text": " Is it possible for there to be information that isn't physically realizable like that isn't instantiated in something physically?"
},
{
"end_time": 6785.026,
"index": 259,
"start_time": 6757.295,
"text": " I mean, that's at least the way we think about stuff. That's that's very much in line with with the I think there's a long tradition of people, physicists have realized this, I think Charlie Bennett being one of them. And I think the you know, most of the founding fathers of quantum information had this bias of thinking that computers are physical entities, they run on physical laws, and therefore they have properties that are"
},
{
"end_time": 6814.684,
"index": 260,
"start_time": 6786.101,
"text": " The idea of information as being an abstract entity that's not embodied in physical systems is really something that somehow doesn't belong to the sort of philosophical viewpoint that we are following."
},
{
"end_time": 6841.152,
"index": 261,
"start_time": 6815.179,
"text": " Um, and ultimately the construct of your information is precisely says that. So it gives you a handle on information that's physical. It says it's a set of states that can be copied. Um, and so that's, you know, when you talk about the set of states of a physical system in dark, you're saying that everything that can, you know, information is actually physical. Okay. Now, Kiara, before we go, I just want to know what's one"
},
{
"end_time": 6870.469,
"index": 262,
"start_time": 6841.578,
"text": " Mistake. What's the best mistake that you've made that has turned out well? Well, I think I think I've made lots of mistakes, but perhaps the the Okay, I think that the one that comes to mind now is this fact that I"
},
{
"end_time": 6899.019,
"index": 263,
"start_time": 6871.186,
"text": " So I think I, I didn't know, I, I, I thought for a long time that I wouldn't be interested in necessarily in, in, in science or physics. So I think my love for physics was, uh, was, came in late in my, in my life. So later, later, let's say then, then say teenage years. Right. So I think I, um, initially thought I would be a writer. So I'm about 10 or something. I had this idea that I would love being a writer and"
},
{
"end_time": 6928.626,
"index": 264,
"start_time": 6899.582,
"text": " I loved languages, I loved literature, poems, everything that had to do with language just fascinated me. And so I ended up sort of working a lot on these things and my own choice in school in back in Italy, I think where you can choose between, it's a bit different from the anglosphere, but I think you have a choice between two kinds of sort of main parts. And one of them is more science inspired, the other one is more"
},
{
"end_time": 6957.517,
"index": 265,
"start_time": 6929.462,
"text": " Classics and literature inspired. So I chose that. And okay, you could think of that as a mistake in a sense that later on, I realized that I really did want to study science. So that happened at the end of the five years of secondary school. But so on the one hand, this meant that I managed to actually just satisfy this love for literature and classics and languages."
},
{
"end_time": 6986.988,
"index": 266,
"start_time": 6958.387,
"text": " And then it also meant that when I switched to science at the university, when I switched to physics, I actually fell in love in the second time in a sense. And that was very exciting. So unlike maybe my other colleagues that somehow had read mostly, you know, they already had had this school mostly within science. So they had being exposed more to say physics and calculus, et cetera. I really just found it then when I was 18 and"
},
{
"end_time": 7016.118,
"index": 267,
"start_time": 6987.807,
"text": " I just, I just found that it was mind blowing and very beautiful. And somehow the fact that I had all of this baggage from the classics side of things made it even more exciting for me to see these things. Uh, because I, it's like seeing them with different eyes and I, so in a sense that was maybe you could classify it as a sort of mistake initially. Uh, maybe I could have immediately started on the parcels of already back in the teenage years."
},
{
"end_time": 7036.152,
"index": 268,
"start_time": 7016.544,
"text": " But I didn't and I'm very happy that I didn't because I somehow it's really nice to be able to see maybe similarities also between different, you know, between humanities and science and physics and physics is really a storytelling. So I ultimately satisfied my life for my love for storytelling, which is actually what I really loved as a child."
},
{
"end_time": 7058.746,
"index": 269,
"start_time": 7036.8,
"text": " by ending up in this field that is really about telling stories about the universe and it's the most accurate, most complete way of telling stories because it really has, you know, you have reality as a checkpoint and you have to sort of make stories are compelling and clear and precise and that's just what makes me fascinated about physics itself."
},
{
"end_time": 7081.613,
"index": 270,
"start_time": 7059.599,
"text": " Yeah, aspect of your book to bring this back that I enjoyed is that and again, the book is the science of can and can't and it's on screen. It's in the description is that you get the sense of a love of writing. Something I'm curious about is many scientists see their exposition. They're explaining to the audience as some necessary evil they have to do at the end in order to get grants."
},
{
"end_time": 7109.957,
"index": 271,
"start_time": 7082.176,
"text": " Because they need to drum up some fervor in the public sphere, but they don't see it as something that they want to do. They like to do the research and speak technically. When you were writing your book, did you find that the writing actually helped you develop your more technical ideas? With TD Early Pay, you get your paycheck up to two business days early, which means you can grab last second movie tickets in 5D Premium Ultra with popcorn."
},
{
"end_time": 7131.152,
"index": 272,
"start_time": 7111.357,
"text": " Absolutely, I think so."
},
{
"end_time": 7158.575,
"index": 273,
"start_time": 7132.09,
"text": " I do find it challenging. So in a sense, maybe, you know, if I followed my least resistance path, I would also sort of tend to just do my own, you know, technique, technical bits of writing and research. But the task of explaining it to someone who doesn't maybe have the same mathematical tools and doesn't maybe even have an interest in this stuff and making it interesting to them."
},
{
"end_time": 7172.807,
"index": 274,
"start_time": 7159.241,
"text": " is a very humbling and at the same time exciting enterprise and I think I really enjoyed it because it clarified in my head also my own understanding of things."
},
{
"end_time": 7200.265,
"index": 275,
"start_time": 7173.148,
"text": " It didn't perhaps lead to new discoveries in the sense, in a strict sense, but I think it's true. I agree with, I think many physicists have said this before, that somehow if you only know a formula and maybe you understand the mathematics, but you can't explain it in simple terms, it means that something is escaping you and you don't understand it yourself. So somehow trying to explain it to someone in simple terms is really a great exercise for everyone to do."
},
{
"end_time": 7226.084,
"index": 276,
"start_time": 7200.862,
"text": " Data for teaching, I think teaching is a similar thing. It's a very nice activity because it allows you to really try to sort of break down your understanding of something very complex into simple steps that you can then explain to people who haven't seen this before. So yeah, definitely, I really enjoyed it. And I really enjoyed writing anyway. So because that's, as I said, this is part of my passions."
},
{
"end_time": 7254.309,
"index": 277,
"start_time": 7227.005,
"text": " And my last question, I know you got to get going. Where do you get your best ideas? When and where do you get your best ideas? Oh, that's very it varies. I think it's really like it's really true that you have to be it's really a creative act. I think it's something that this probably is true for everyone is doing some creative work. You have to be relaxed."
},
{
"end_time": 7265.811,
"index": 278,
"start_time": 7254.889,
"text": " You have to be free of worries if possible. So somehow you have to be able to Bring yourself to a sub space where you're not concerned about, you know daily problems"
},
{
"end_time": 7293.66,
"index": 279,
"start_time": 7266.22,
"text": " And you have to be, at least as far as I'm concerned, you have to feel free to explore things without constraints. So I was very lucky in my PhD or the field because all my supervisors, I mean, David was a collaborator. He's great at this, but I think also R2 Records was my supervisor at the time. They are masters of this freedom seeking attitude. So I think I was"
},
{
"end_time": 7321.527,
"index": 280,
"start_time": 7294.053,
"text": " really inspired to be just so I was let let you know left on my own and to just think and I think that's the best state you can be in so you're not forced by someone to say okay you have to work on this problem or this other problem you have to be free to let your mind roam and and and explore things like you are you know as Newton used to say right that you're on a seashore and you're sort of"
},
{
"end_time": 7347.466,
"index": 281,
"start_time": 7321.954,
"text": " looking at pebbles and you're like a child playing with things. You have to be in this fun seeking free state to have your best ideas. That's as far as I can tell. And this is, as I said, a constant also, actually, my mentor, Mario Rossetti, who was the person, by the way, who introduced me to quantum information back in my masters in Torino. I think all of them, all of my mentors had this feature."
},
{
"end_time": 7376.271,
"index": 282,
"start_time": 7347.978,
"text": " And they all said that their own mentors were like that. So I think somehow it must be a constant. It's true of, I would say it's true of most creative activities that you need to be in that state to be inspired. And you have to be able to sit quietly somewhere to be inspired. So it could be that you are on a journey. So sometimes I enjoy thinking when I'm traveling, it doesn't have to be that you're in a super quiet place, but somehow you're tranquil. And so, you know, you're in your own zone and then you can think."
},
{
"end_time": 7405.691,
"index": 283,
"start_time": 7376.698,
"text": " Um, and, and so the, the place can vary. It could be on a walk. It could be on a swim. It could be while I'm traveling, uh, or I'm sitting at the desk, but there has to be a moment where you're sitting there in this space and waiting for inspiration. And sometimes it doesn't come, you just have to be patient and wait for it. But when it comes, then it's really nice to follow it. And, and yeah, it's, it's really very much like an artist. That's, that's, I think the way I work and presumably all my colleagues also do."
},
{
"end_time": 7427.773,
"index": 284,
"start_time": 7406.374,
"text": " You mentioned you're waiting for the inspiration or just the inspiration occurs to you? Like, are you sitting here like, come on, inspiration, come on. Yeah, I think I think sometimes you have to be patient and wait is a bit like when you're sitting to, you know, if you're in outside walking and you want to see some rare animal in the wood, you just have to be patient. But you've got to go there."
},
{
"end_time": 7454.94,
"index": 285,
"start_time": 7428.575,
"text": " And I think that's the thought that's maybe the hard thing to do. You have to cultivate these things, especially if you're not, you know, there are also things that go in the way I think of research these days, not just daily problems with, you know, everyday life, but I think also, you know, grant applications, as you said, and admin duties, etc. They're all enemies of creativity, optimizing research oriented ideas. So I think you really have to find"
},
{
"end_time": 7482.056,
"index": 286,
"start_time": 7455.555,
"text": " some way to guard your time and say also no to things and find some free time to wait for inspiration. And you've got to be somewhere for it to visit. So in that sense, yeah, I think I very much think of me myself as sort of being one of those explorers that are waiting for to see a rare animal in the forest. I think that's how I think of myself."
},
{
"end_time": 7493.797,
"index": 287,
"start_time": 7482.5,
"text": " Thank you, Chiara. Thank you for spending so much time with me and the audience. Thanks to you. Yeah, it was great. Great being here. Thank you very much for the questions and for listening. Yeah. Thank you. Take care. Bye. Bye."
},
{
"end_time": 7518.575,
"index": 288,
"start_time": 7494.514,
"text": " The podcast is now concluded. Thank you for watching. If you haven't subscribed or clicked that like button, now would be a great time to do so as each subscribe and like helps YouTube push this content to more people. You should also know that there's a remarkably active Discord and subreddit for theories of everything where people explicate toes, disagree respectfully about theories and build as a community our own toes."
},
{
"end_time": 7536.613,
"index": 289,
"start_time": 7518.575,
"text": " Links to both are in the description. Also, I recently found out that external links count plenty toward the algorithm, which means that when you share on Twitter, on Facebook, on Reddit, etc., it shows YouTube that people are talking about this outside of YouTube, which in turn greatly aids the distribution on YouTube as well."
},
{
"end_time": 7557.381,
"index": 290,
"start_time": 7536.613,
"text": " Last but not least, you should know that this podcast is on iTunes, it's on Spotify, it's on every one of the audio platforms. Just type in theories of everything and you'll find it. Often I gain from re-watching lectures and podcasts and I read that in the comments, hey, toll listeners also gain from replaying. So how about instead re-listening on those platforms?"
},
{
"end_time": 7586.647,
"index": 291,
"start_time": 7557.381,
"text": " iTunes, Spotify, Google Podcasts, whichever podcast catcher you use. If you'd like to support more conversations like this, then do consider visiting patreon.com slash Kurt Jaimungal and donating with whatever you like. Again, it's support from the sponsors and you that allow me to work on toe full time. You get early access to ad free audio episodes there as well. For instance, this episode was released a few days earlier. Every dollar helps far more than you think. Either way, your viewership is generosity enough."
}
]
}No transcript available.