Podcasts

Chiara Marletto | The Science of Can & Can’t

about the episode

In the Existential Hope-podcast series, we invite scientists to speak about long-termism. Each month, we drop a podcast episode where we interview a visionary scientist to discuss the science and technology that can accelerate humanity towards desirable outcomes.

Chiara Marletto is a physicist at Oxford University, and author of the book “The Science of Can and Can’t”, where she writes about the concept of counterfactuals and constructor theory.

This is a new and different approach to physics where the idea is that the physics most of us are familiar with is mirrored in a way that is very limited. There is a large class of things that science has neglected, and according to this theory, there are questions of what is and isn’t possible, which could provide scientific explanations to many questions in the long run. Those unwilling to bring this concept to science prevent us from making progress on certain fundamental problems.

In this episode Chiara talks about (among other things):

  • Constructor theory
  • Her vision of existential hope for the future
  • Risks for the long-term future
  • The best piece of advice she ever received

Transcript

The transcript below follows a question and answer format, highlighting the X-hope podcast special. Allison Duettman and Beatrice Erkers of the Foresight Institute sit down to co moderate a session with guest Chiara Marletto. The podcast aims to ask guests what they are focusing on in terms of their professional career, what motivated them to reach the present point, what hopes or risks they see for the future, as well as what advice they may have for fellow listeners. In this podcast, Chiara Marletto, a physicist and author of “The Science of Can and Can’t,” sits down with us, as a senior Foresight fellow, and shares her thoughts.

Allison Duettmann: Hello everyone! Welcome to Foresight’s existential hope group. Our podcast is a new special series we started this year, taking a few folks that most inspired us and interviewing them. Basically, we brush upon what they are currently working on, what it is that makes them most optimistic about their future, and the path they envision for society as a whole. We already had our inaugural podcast for this series with Christine Peterson, the co founder of Foresight. Many of you were there for the recording of it. Also, many of you have noticed that an interesting bit of our podcast includes being more action- oriented. For instance, we ask for a day in the life to inspire others to think more concretely about this possible future. On the back end, we place it on the existential hope website and try to call for artwork regarding their specific answer. In Peterson’s case, our call for artwork was for a future where longevity is possible to revive a dog through cryonics. These exercises are yet another way to make these hopes and efforts for the future seem more concrete. You can also apply by writing a story of a day in the life when a dog gets revived. So we are trying to make this process a bit more interactive as we imagine the future in a more concrete way.

Shifting gears for today’s podcast, we look at a highly new, but deeply optimistic, approach to the future. It is much more theoretical, but has huge implications for how we interact with the universe in the long run. We are very fortunate to have Chiara Marletto here. Chiara, I have to say since your last podcast with us, which was some time ago, it was one of the most watched podcasts we have had in this series. You have published a fantastic book since then. Beatrice, would you want to give us a better outlook in regards to what we can expect for today?

Beatrice Erkers: Sure, so thank you for joining us Allison and Chiara. A brief introduction is that Chiara is a physicist from Oxford, as well as a Foresight senior fellow. We are especially glad to have her here today. You wrote this book called, “The Science of Can and Can’t” which has been quite inspiring, and I for one have been reading it recently. It is fun to discuss in this group. To dive in, the book is this concept of counterfactuals, constructor theories as I believe it is called. It is this new, different approach to physics that you have been a part of working on. It is this idea that the physics most of us are familiar with is mirrored in a way that it is limited. There is a large class of things that science has neglected, so according to this theory, there are questions of what is and isn’t possible, which could provide scientific explanations to many questions in the long run.

Those unwilling to bring this concept to science prevent us from making progress on certain fundamental problems. I believe the way it fits in this group is that it is about thinking what is possible versus what isn’t, which is what we are thinking in this group with what possible futures are out there. This, of course, is not easy. Nevertheless, it is important because so many things are possible, even if we do not know it yet. That is why we are especially excited to have Chiara here in this group. So, I hope you can tell how enthusiastic we are! Let’s dive in.

Allison Duettmann: Perhaps you can bring us up to speed on what got you started on your current track? What do you currently work on? You can coat a bit of what we tried to outline. You have a few fantastic examples.

Chiara Marletto: My background is theoretical physics and I am in the field of quantum information theory. This field is where we study how computers can run on physical systems that are not ruled by the standard laws of Newtonian physics, but instead quantum theory, which is the current best explanation we have for the microscopic world. It has some very interesting effects, which are very important for speeding up computation. That is how the field of quantum information theory came about and that is where I did my PhD. While I was completing my degree, I stumbled upon constructor theory. I went to a talk that David gave on the topic where he laid the foundations of the new idea down. It was basically to go beyond the theory of quantum computation to better cover computers and new machines called constructors.

Constructors are objects that you can program to perform many tasks. The great mathematician John von Nuemann thought of a machine that should be the ultimate generalization of the computers we have now, which he called the universal constructor. This machine is programmable to perform any tasks or transformations physically permitted. You can think of it as sort of an all powerful 3D printer. The machine hasn't been realized yet. Since von Nuemann mentioned the general idea, no one has really developed a theory of the machine to proceed. Nevertheless, David thought now is the right time to try and tackle this problem of having a general theory for the machine, which is based on the conceptual ideas we developed in the meantime in the quantum theory of computation.

As such, I joined in and tried to twist this new theory as a new way for formulating the laws of physics as well. Not just formulating the foundations in a way for a new kind of technology in the very distant future, which is this idea of universal constructors, but also for a new kind of physics based on putting constraints on what is physically possible or impossible. These constraints are ultimately the very things that make the universal constructor work by providing a theory as a constructor. However, they also provide general principles of physics, like the conservation of energy, or the second law of thermodynamics. Yet, more general than that, they can constrain future fundamental theories. That is essentially the overall vision of what I am doing and how I got to that point.

Allison Duettmann: Wonderful, that is incredibly far reaching. Could you maybe pinpoint a few of the applications that could potentially be affected by this? Or are there any practical implications this theory could have, now or in the long run?

Chiara Marletto: Yes, I think the universal constructor is the ultimate goal, far in the future that we can think of. Along the way, there are a number of promising things we have been developing. I’ll start with some more foundational ones that deal with physics, building more towards technological ones. Firstly, physics is generally where we come up with ways to explain how the fundamental building blocks of the universe work. This may seem very far from application, or not very interesting, but actually, once we learn of these new laws of physics, we also learn new limitations for machines we can build. If you consider how humanity made progress since the beginning, even before physics when we discovered something useful like fire or the wheel, there was a mechanism associated with such a discovery. We had an implicit theory of how that object worked. For instance with fire, we knew how to preserve it and start it and so forth. Later on, this became more explicit, say with the heat engine which was branded with thermodynamics. As a result, when there was a development with physics, there was a parallel with technology. Now, when we are developing a new application or theory, we could expect to see many more applications around that new idea.

In the case of constructor theory, we are trying to formulate more general principles which can describe how new information works. This is very important for theoretical studies because we know how to manipulate information, but until David and I worked on this, we didn’t know what are the regularities in physics that allowed for information. We took for granted having the information and that’s it. However, with these new principles of constructor theories we developed, we can now say we have some laws that allow for information to be possible in our physical world. We can say this calls for some transformations being possible, specifically copy-like transformations and others like permutations. It is interesting because once we understand what it is in the laws of physics that allows them to support the idea of information, we can then start thinking these are regularities that we would like to have in the future of physics, beyond the current ones. We expect the fundamental laws of physics, including the ones now like quantum theory and general relativity, to be modified in the future because they clash with each other and we will need a better theory that supersedes both. It could be very useful to act as guidelines for what new theories could be. Then we can go in the more applied direction, like with thermodynamics.

A more traditional physicist would say thermodynamics is very useful to understanding heat engines and machines that can power cars, airplanes, and so forth. Nevertheless, it is not necessarily a fundamental branch of physics. In the field we have an hierarchy of how close to the fabric of reality theories are. Thermodynamics is not that close. This is a bit worrying because on the other hand, thermodynamics introduces an irreversibility in the world. Irreversibility is something we use all the time in machines and applications. The fact that this is considered an imagined phenomenon, not as important to physics, sounds a bit at odds with our experience. Constructor theory allows us to develop a new version of thermodynamics where we can extend the laws we currently have so they apply to those scales. So that is another very interesting direction. Developing this would allow us to have a more general theory of thermodynamics, which not only applies to the microscopic world as we currently do, but it would also allow us to develop a more general set of laws that apply to objects at every scale, even microscopic ones.

The third direction is the physics of life. This is a bit closer to the universal constructor because von Nuemann wanted to find a way to mimic, in all respects, a living system when developing the universal constructor. He noticed the model Turing machine, which was the basis of our computer, could be programmed to do a lot of things, but not to create a replica of itself. Nevertheless, the universal constructor can replicate itself, which is what cells do in the physical world. So, understanding the fundamental limitations of a self-reproducing machine is a really interesting question. I do not think von Nuemann quite answered it because he studied it in a more artificial domain that is not exact to the real world. As such, we want to construct a theory of physical limitations of self-reproducing machines, which is interesting because it will give us a way to understand the basic limitations of our technology when trying to emulate properties of life. This would mean a number of things, such as extending life, making it more powerful, resilient, and so forth, which is really exciting to me. These things lead us to some applications of this new theory.

Allison Duettmann: They are all amazing. The third one is incredibly inspiring. Do you have any specific research strengths that someone should look into if they are new to the field? Any questions that keep you up at night?

Chiara Marletto: A lot! I think, specifically with the last direction, we like to talk about regularities of certain entities in physics. For instance, in thermodynamics you can say it is about energy and how it can be transferred from one thing to another. With self-reproducing machines, we do not quite know what is the substance they run on. That is one interesting question. We have hints at what that is by knowing biology and how biological systems have come up through natural selection and evolution. In this process of creating replicas more resilient, there is some resilient information in the genes, which can power self-reproduction as well. If you think of other complex resilience in the universe, these objects usually contain this sort of resilient and capable information for causing transformations in the environment. Now, for a physicist this is quite exciting because there are a lot of questions you can ask regarding the entity of knowledge. For example, is there a law that says how knowledge can be created? Under what conditions can it be preserved or grow more resilient? These questions now are only formulated at the level of informal questioning, but I think the hope is to find a new branch of physics that can answer these questions in a quantum way, the same way we do for entropy, energy, and thermodynamics. You know, if someone were to enter the field in this stage, that is a very interesting question to look into right now.

Allison Duettmann: Of course. As we saw with your last podcast reaching so many views, it seems as though your work has gained some traction. How have you seen the book being received? Has it already spun off new research or investigation? At least, as we can tell, it has been quite successful in getting this abstract and powerful idea across.

Chiara Marletto: That is a very good question. When you write a book, it goes out in the wild on its own and you lose touch with it in some sense. It is sort of an example of a self-reproducing machine, generating more ideas out there. As far as I can tell, I have had interesting feedback. Scientifically speaking in particular, the idea that was new and created the most interesting comments was the fact that you could somehow do without these standard machinery of laws of motion. These laws are usually considered the most fundamental thing in physics, while still being capable of making useful predictions that you can still test. We sort of knew this in some sense as we have used principles in the past to make predictions, such as with thermodynamics. However, it had not been spelled out in such a general way, as I did in the book. I think that somehow resonated with a lot of people. There are some interesting experiments we are trying to devise based on predictions that came from the principles for constructor theory. This is a very interesting endeavor because it highlights how you could make a testable prediction from these general principles.

Allison Duettmann: That is much more than I could’ve hoped for. That is great. I will check more into this as it continues. Alright, I will hand it over to Beatrice to get into the more X-hope-based questions of our podcast.

Beatrice Erkers: Yes, thank you. Since this is the existential hope group, I have some specific questions on the topic of existential hope. Feel free to get creative with these questions as they are meant to think widely and explore potential possibilities and opportunities. I will start with asking, what is a vision of existential hope, as opposed to existential angst for the long term future?

Chiara Marletto: Right. Some of the things I am going to say are in line with what David Deutsch has been advocating. I find a view of hope for the future in the fact that it is possible to solve problems. When you start life, you have a natural tendency to avoid problems. As you grow up, you realize it is not quite possible because problems inevitably arise. However, the good thing is that under ideal conditions, it is always possible to find solutions to problems. That is very hopeful. If you see each setback in your scientific research as something negative, you can get depressed, or even give up. If you consider these opportunities to learn more about your task and find the right way forward, that is much more helpful. For instance, Karl Popper has this view that while we cannot avoid problems or mistakes, it is quite desirable as it is the first step towards progress and improvement. It is rooted a bit in constructor theory, as well. This opposition becoming possible is quite hopeful in itself because as long as there is no law in physics saying it is impossible, the rest is possible. That realization in itself is a hopeful message stemming from the foundations of physics.

Beatrice Erkers: That is quite hopeful. We seem to underestimate the power of problems in a way. In thinking of solving problems, is there one particular breakthrough that could happen in the next five years that would validate we are on track to a brighter future?

Chiara Marletto: This is a very broad question, but if I go along with this idea of problem solving being important I can think about a few things. One thing we haven’t cracked yet, and I am not sure if we can reach it in five years, is finding ways of understanding how to program a machine to do the same. We have the AI field, but I don’t think this is enough, for those are more task-oriented algorithms. What I have in mind is more so a computer algorithm that allows the computer to be as creative as the brain. If you think from a foundational point of view, we know human brains have the capacity to solve problems. However, it is not possible on demand, and it is quite hard to do it in a reliable way due to many variables. One way to work towards this development for machines would be to find a theoretical underpinning to the process of knowledge. It would be hopefully to see that we are moving in that direction. To this day, we have yet to fully emulate the human being or understand from the foundational point-of-view what it means to create knowledge. We need to crack that as humanity, for it will allow us to solve more problems.

Beatrice Erkers: Are there any particular risks you see moving forward that we seem to undervalue?

Chiara Marletto: There are many risks, more than we can imagine sometimes. We have become comfortable with the tools that we had, so we are using them a lot for research. However, perhaps we are not looking hard enough for generally new, disruptive tools. For instance, ideas that are different to the ones we have used for a long time come with some hesitancy. It would be nice if we had more of these attempts. It seems to me we are not so comfortable with failure, so the idea of doing something weird or different from the common brings great possibilities. We should try a lot of things, comfortable with the idea that many of them could fail, understanding it is necessary to potentially find a gem overall. This is especially important for young people and a new generation of scientists to not just try out familiar things. That could be a huge risk. We should be more ambitious, within reason of course, to develop newer and stronger ideas. I feel it is underestimated, but we need to take it seriously.

Beatrice Erkers: I appreciate that. It is quite in line with Foresight’s mission that we are trying to get attention, funds, and resources for technologies considered a bit out there to many. Would you describe yourself as optimistic about the long term future? If you are, what would you say made you so?

Chiara Marletto: Yes, I think I am naturally optimistic in the sense that I have this science background. I think as humans, we are able to solve a lot of problems. It is not just optimism about my own life trajectory, but also for humanity in general. We can do so much. I am unsure about the actual trajectory we will follow to get there, of course. I am aware that various situations can develop and cause setbacks, such as the pandemic we have been facing. However, I do feel that we have generated a lot of knowledge over the past few centuries, and it seems to me we just have to keep trying. We have to be wary that we do not stick by the mindset that we have picked all of the low hanging fruit and become stagnant. Instead, we have to climb further up. As long as we stay on track with problem solving and idea formation, we have a good chance to develop our trajectory towards better possibilities in the future.

Beatrice Erkers: Definitely. Steering the ship in the right direction so-to-speak. We need to start thinking about the positive counterfactuals of a beautiful, abundant future. It often seems it is very hard for people to imagine these utopias. Why do you think it is hard for people to envision these positive scenarios for the long term future? Do you think there is anything we can do to change that?

Chiara Marletto: On one hand, it is natural for us to be apprehensive of change. It is a mechanism of self-preservation when we think towards gloomy scenarios to be better prepared for those, which I do not necessarily think is a bad attitude. However, if we concentrate on that too much, then we lose the creativity to remember the good scenarios as well. We should be able to rely on our ability to solve problems and focus on those just as much as the gloomy scenarios. Also, if we focus too much on these negative possibilities it creates stagnation, which prevents us from trying and ultimately moving forward. Certainly, I believe that at least entertaining the same amount of good scenarios as bad scenarios is healthy moving forward.

Allison Duettmann: Wonderful. I’ll jump back in here. This is a nice segue into touching upon active and action-oriented parts of our podcast. You have a very beautiful mind. I remember reading through Infinity, around chapter 8 or 9, which was deeply optimistic about the future. It is sort of how you mentioned trying to make progress on anything not specifically stated as impossible in physics. Taking it a bit further, we try to concretize this thinking. There is a term by Toby Ord taken by Tolkein, known as eucatastrophe. We currently have a bounty out there to create a better term for this. Essentially, it means the opposite of a catastrophe, in which something of better value comes forth after an event. Do you happen to have a term for what you would consider a deeply optimistic event?

Chiara Marletto: I actually like that term, I think it is a good one.

Allison Duettmann: Great, we will go with that term then for a second. Could you perhaps envision a eucatastrophe moment for us? Almost like a day in your life where something happens and afterwards you would be much more optimistic about the long term future. Usually, we aim for these events to allow people to envision themselves in that specific future where such a positive event occured. We want to create a prompt around this so people can write out a story, or create a piece of art, to help concretize such a possibility even more so. Does anything come to mind?

Chiara Marletto: There is an event that brings together a few of these things mentioned so far. It may more so be a transition that we would go through as humanity. My view is that humans can think in a way that allows for creating ideas and solutions. Thinking of our humanity and survival, we should achieve a state where more people are able to just think freely and not worry about day-to-day survival problems. I am imagining an ideal situation where we will reach a time in our lives to think about various problems in peace, unbothered about pressing, survival issues. One event which could improve these chances for revival would be to have a eucatastrophe where everyone can divert their mind. Where they are not worried about paying for rent or eating and just being free to think.

Currently, our technology is supporting us very nicely to where we are moving towards that possibility. This type of event should be coupled with the event that people can grow up to naturally think in a creative way. How do we instill this love for knowledge and creativity? Schools aren’t really doing a good job for that. They seem to be repelling people from knowledge versus getting them into it. So, that would be another eucatastrophe for getting people back into problem-solving, knowledge, and curiosity overall.

Allison Duettmann: I love that. It is very concrete. I think that there was one video from David regarding why flowers are beautiful. That was deeply inspiring for me, sparking wonder and excitement. If you look at how many school systems operate, you couldn’t come up with a better way to keep people from coming up with ideas intrinsically. I recommend people to search for this video. Also, are you at all interested in the taking children seriously paradigm? Or is that unrelated?

Chiara Marletto: That is definitely related. I have a long-lasting interest in this issue regarding how to make a better education system possible in some sense. There are various proposals. Certainly, the fact of taking children seriously presents that you put the idea of fun or interests at the center of education. I am unsure how this can be realized on a large scale, but I do think it is very important to revise what we do in schools. We should be trying to inspire people as opposed to ticking boxes for the next exam. I was very lucky with some teachers that were not like the mainstream. I think we should do that more. It is very important.

Allison Duettmann: I agree. This fits nicely with the point you made earlier regarding incentives for science to be creative and not scared to make mistakes. We will try to take a glimpse of this optimistic vision you described. As we are wrapping up here, on the track of existential hope: Was there a specific eucatastrophic moment in your own life that changed, or cemented, the course of your passions? It would be interesting to see if there was a profound moment or piece of advice you received that truly helped you out.

Chiara Marletto: Yes to both questions. In terms of an event that shaped me, I remember a continuum of things. Nevertheless, up to a certain age, I think 18 years old, I was into humanities and philosophies. In Italy, things are science-oriented or humanities-oriented in education, and I initially chose the latter. In the end, through various interactions with teachers and my parents, I felt as though I was missing something. I thought it was a bit crazy to change my trajectory because I did enjoy it, but I did not want to miss the rest of the knowledge available to me. I remember a philosophy teacher who mentioned Popper in her lecture and quantum physics. I asked her what she meant regarding a certain remark, and she apologized saying it is not her field and she didn’t know. That stuck with me because I didn’t want to be like that and I wanted to know.

Very gradually, this seed started to bloom and I chose to do physics at the university I went to. This was a huge gamble, because I was not sure at all, but I loved it. So it was sort of a eucatastrophe. Also, a piece of advice that was important to me and my journey was actually given to me by various people, indirectly and directly. Parents, teachers, mentors, David, and so forth have mentioned doing things for fun. You know, just follow what is fun in whatever you are focused on. The idea is that it is the best bet, because even if you do not succeed, you enjoy the path along the way nonetheless. If you choose something because you want a good career or prestige and then you don’t succeed, that is just unfortunate on both fronts. If you don’t put your mind and your heart into it, it is very hard to achieve something regardless.

Allison Duettmann: I wholeheartedly agree. I love what you said about Popper. He was the founder of my department at LSE. Alright, we are out of time, and I cannot thank you enough. Thank you for all of this existential hope. I certainly feel energized for what can be possible. You have an incredibly beautiful mind. I encourage everyone to go out there and buy “The Science of Can and Can’t”. Also, thanks to all of you who joined us.

Read