The Joe Walker Podcast - In The Foothills Of A Pandemic - Yaneer Bar-Yam

Episode Date: March 9, 2020

Yaneer Bar-Yam is a physicist and the founding president of the New England Complex Systems Institute.Show notesSelected links •Follow Yaneer: Website | Twitter •Endcoronavirus.org •Dynamics of Complex Systems, by Yaneer Bar-Yam •Making Things Work, by Yaneer Bar-Yam •'Coronavirus Disease (COVID-19)', research by Max Roser and Hannah Ritchie •Powers of Ten YouTube video •'An Introduction to Complex Systems Science and Its Applications', 2019 paper by Alexander Siegenfeld and Yaneer Bar-Yam •'The Architecture of Complexity', 1962 paper by Herbert Simon •'Science and Complexity', 1948 paper by Warren Weaver •'More is Different', 1972 paper by Phil Anderson •Scale, by Geoffrey West •Johns Hopkins University coronavirus interactive map •'Systemic Risk of Pandemic via Novel Pathogens -- Coronavirus: Note', January 2020 note by Nassim Taleb, Yaneer Bar-Yam, and Joe Norman •The Square And The Tower, by Niall Ferguson •'Long-range interaction and evolutionary stability in a predator-prey system', 2006 paper by Erik Rauch and Yaneer Bar-Yam •'Transition to Extinction', 2016 article by Yaneer Bar-Yam •'Nonpharmaceutical Interventions Implemented by US Cities During the 1918-1919 Influenza Pandemic', 2007 paper by Howard Markel et al •'Strategies for mitigating an influenza pandemic', 2006 paper by Neil Ferguson et al •Join the fellowship of the doers: [e-mail] discussed •Yaneer's background and parents. 11:39 •Powers of Ten. 12:06 •Highlights from Yaneer's time as an MIT student. 15:29 •The role of chance in our lives. 21:28 •What is "complexity"? 25:42 •Complex systems. 30:23 •Emergence. 37:06 •Phase transitions. 44:26 •Self-organization. 49:48 •Universality. 55:12 •Applying complex systems science to the Arab Spring. 1:03:13 •Taking stock of the coronavirus epidemic. 1:12:47 •What is the current best estimate for th...See omnystudio.com/listener for privacy information.

Ladies and gentlemen, welcome back to the show. This episode of the Jolly Swagman podcast is brought to you by Freelancer.com. Are you trying to start a business or do you already have one that needs extra help? You're doing yourself a serious disservice if you're not using Freelancer.com. It's the world's largest crowdsourcing marketplace and that's both by number of users and projects posted. Full disclaimer, I know the CEO, Matt Barry. He's a great guy, a very uncorrelated thinker, and Matt has produced a true Aussie startup success story. Freelancer.com is where you go if you want to start a business

or take an existing one to the next level. It does this by connecting entrepreneurs, that's you, with other entrepreneurs around the world, the freelancers. So if you need help with getting a website built, a mobile app developed, graphics designed, or even something more complicated like financial research, freelancer.com is where you can go to hire skilled professionals inexpensively. It can be for almost anything. It's home to over 1600 different categories of work. It's free to post a project and to get quotes, so there's really no reason not to try it out.

But in addition to that, listeners of this podcast can go to freelancer.com slash swagman, where you can get an expert from their team of recruiters to reach out and help you find the best freelancer for your job and budget for free. So follow the link freelancer.com slash and budget for free. So follow the link freelancer.com slash swagman, sign up and select the recruiter upgrade. I tried it out a couple of weeks ago and put up an ad for animated videos for the podcast. I got about 13 bids from real

people around the world within about 20 minutes. No job is too big or too small. Go to freelancer.com slash swagman and post your project today to turn your dreams into reality. Who doesn't want that? This episode is also brought to you by Blinkist. Blinkist is an app that condenses the key takeaways from the best non-fiction books in the world into 15-minute blinks which you can read or listen to. It might have seemed like Blinkist and I were always meant to be. I do try to read a lot of non-fiction, but the company has a checkered history with me. I remember the first time I heard about them. It was 2016. I was in a gym on Smith Street in Darwin, and I listened to

my first Blink. After that, I thought, nah, not for me. No one shall do my reading for me. I'm a purist. Then in 2019, the company contacted me about sponsoring the podcast. I negotiated a great deal, but then turned them down for the same reason. No one should do your reading for you. There are no shortcuts. That was July, 2019. What happened next was I started using the free annual subscription the company gave me during the negotiation process. And I changed my mind. I worked out how to hack Blinkist. You see, I was thinking about it all wrong. Blinkist isn't a substitute for books. It helps you decide which books to read because the cost in reading, as one of my podcast guests once reminded me, isn't the sticker price

on the book, but the hours of opportunity cost spent reading it. And you need to know before going in whether it's worth your time. Imagine you have a really smart friend who's read all the great non-fiction books in the world and before you commit the time and money to reading a book you check with your friend, is this worth reading? They give you an intelligent but concise summary of the whole thing, while Blinkist is that friend. It's kind of like the Amazon Look Inside feature or the Kindle sample feature, but it lets you understand the whole book rather than just reading the first chapter. Of course, there's no substitute for reading the actual book and understanding the author's reasoning in all of its detail, but in deciding what to read, don't judge a book by its

cover. Use Blinkist. Go to www.blinkist.com slash swagman, where you can get 25% off an annual subscription and you get to try Blinkist Premium for free for seven days. Don't be a dummy. Get on it. That's www.blinkist.com. You're listening to the Jolly Swagman Podcast. Here's your host, Joe Walker. Ladies and gentlemen, boys and girls, swagmen and swagettes, welcome back to the show. This episode is undoubtedly the most important I've ever released in terms of the public interest. Our guest is Yenir Bayam, MIT educated physicist, one of the founding fathers of the science of complex systems, and founding president of the New England Complex Systems Institute.

The episode was recorded on Sunday the 8th of March 2020 in Sydney. With no hint of exaggeration or hysteria, the coronavirus epidemic looks likely to be one of the defining catastrophes of our lifetime. Unless we take drastic measures to stop it in its tracks. To appreciate why the risk is so grave, the first and most important thing you need to do is to viscerally understand the explosive power of exponential growth. Exponential growth does not come naturally to our thinking. Test your intuition with the following parable, related by the Islamic scholar Ibn Kalikhan in 1256. It's known as the wheat and chessboard problem. If you placed a single grain of wheat upon the first square of a chessboard, two grains on the second square, four grains on the third square,

and so on, doubling the number of grains with each successive square, how many grains of wheat would be sitting on the chessboard at the very end? In one version of this story, the inventor of chess asks his king for a payment of wheat equal to this amount. The king laughs. This seems like a paltry prize for the creation of chess, and so he agrees to pay the inventor by this method. Until the king's treasurer informs him that the sum would outstrip all of his resources. The inventor understands something that the king doesn't, that is, geometric sequences. On the entire chessboard at the finish, there would be roughly 18.44 quintillion grains of wheat, weighing more than 1 trillion metric tons, more than 1600 times the global production of wheat today.

Exponential growth is counterintuitive. And that's why most people aren't psychologically prepared for the accelerating spread of this novel coronavirus. And note the word novel. We have no immunity to this. One simple metric for understanding the spread of a virus is the doubling time. That is, the time it takes for a number of cases to double. If we have 10 cases today and 20 cases in five days time, the doubling time is five days. So what's the doubling time for coronavirus? Max Rosa and Hannah Ritchie at ourworldindata.org recently looked at the Global World Health Organization data up to and including the 7th of March 2020. They found the following. The doubling time for the global number of cases including China was 21 days. The doubling time

for the global number of cases excluding China was four days. The fact that the doubling time is longer when China is included is of course due to the fact that the doubling time is longer when China is included is of course due to the fact that the number of daily cases has declined in China since lockdown. This has happened very quickly. On the 29th of December 2019, Chinese authorities first identified a cluster of similar cases of pneumonia in the city of Wuhan in China. By January 30, there were 9,776 confirmed cases of coronavirus worldwide, and there are now more than 100,000. Extrapolate that forward, not linearly, but exponentially.

So what should you expect if you catch coronavirus? Well, the World Health Organization described the cases in China up to the 20th of February 2020, and they found the following symptoms. The most common symptoms are fever and a dry cough. Close to 90% of cases had a fever. About two-thirds had a dry cough. The third most common symptom was fatigue. Almost 40% of cases suffered from that. The average incubation period is 5 to 6 days, although that varies within a wide range of between 1 to 14 days. About 80% of cases are mild. On average, the disease for these cases lasts 2 weeks. Again, this is based on the data from China. The rest are more severe. For severe and critical cases, the disease lasts three to six weeks,

and the World Health Organization reports that about a quarter of severe and critical cases require mechanical ventilation. Estimates as to the fatality rate vary, but I'm not sure that's so important. Think about it this way. Star Harvard epidemiologist Mark Lipsitch has recently stated that his base case is that 40 to 70% of the global adult population will be infected in the absence of strong countermeasures. If you scale up even a 1% fatality rate against a group so large, you start to see the problem. And this is to say nothing of the second and third order consequences for strained healthcare systems, nor is it to say nothing of the second and third order consequences for strained healthcare systems, nor is it to say anything of the attendant social and economic disruption.

It's important to note that the coronavirus does not represent an existential threat to our species. We will get through this. But we are, for the moment, to bastardise from Henry Kissinger, in the foothills of a pandemic. For a deep understanding of the structure of pandemics and how we can take the fight to coronavirus, this conversation with Yanir Bayam is as good a place as any. In the first half of this episode, you're going to get a huge download of information. We talk about the science of complex systems. You're essentially going to be getting a one-hour crash course from one of the guys who literally wrote the book, hopefully spiced up with a bit of questioning from yours truly. Then we're going

to apply this intellectual toolkit to the problem of COVID-19. If the first half of the conversation is too academic for your tastes, I implore you to try and listen through anyway or to skip ahead to the stuff on coronavirus because it's really, really important. If you do want to skip straight to the coronavirus, that conversation begins about one hour and 10 minutes into the interview. My hope is that after listening to this podcast, apart from developing a better understanding of complex systems, people will stop their rationality signaling of poo-pooing concern about the virus by saying it's just another flu, and more people will take actions to mitigate its spread. Without much further ado, please enjoy this conversation with Yanir Bayam. Yanir Bayam, thank you so much for joining me.

It's a pleasure. I'm both exhausted and exhilarated. Exhausted because we organized to do this about two days ago, and I've been up all night preparing. But I'm exhilarated because I think this might be the most important podcast episode I've recorded so far. First, we're going to go through a crash course in complexity science. You are, after all, one of the founding fathers of the field. And then secondly, we're going to apply those insights we'll learn to the coronavirus epidemic as it remains at this moment. But before we begin, I'd like to introduce you to our audience and talk about your background. Younir Bayam, born in 1959, physicist and president and founder of the New England Complex Systems Institute. Younir, where were you born?

In Boston, actually. What was your mother Miriam's profession? A developmental psychologist. She was born in Jerusalem. Right. And what was your father Zvi's profession? He's a physicist, so I'm a physicist's son of a physicist. He is an experimental high-energy particle physicist.

Many years ago, your father told you that he used to imagine the movie Powers of 10 when he was a child under difficult circumstances. What is the movie Powers of 10 with a view of, say, someone lying down, and then to zoom away and to see the scene from farther and farther away, say from where a balloon might look down, and then from where a spaceship in orbit around the Earth would look down, and then maybe from the moon, and then further and further away. Of course, you see the scene recedes, it becomes smaller and smaller, and then you see the Earth and you see the solar system and you see the, eventually you see the galaxy and you see from very far away. So the point is that you see yourself in the context of the larger space that we live in.

And in an important sense, we realize how tiny we are and how insignificant we are. And then it does the opposite where it zooms in and closer and closer and goes down to the molecular level. And of course, compared to the molecular level, we're huge. Now, for my father, he was a child in Europe during World War Two and was escaping from Poland and the Nazis. And the situation was tremendously difficult. And it's hard to understand the world in that context and to cope with it. But by thinking about this as he went to sleep, I'm sure that it provided some perspective on everything that was going on as being maybe not as important as it seemed if you were embedded in it.

So that was before the movie. The movie came out in the 1970s. So was he thinking about the book Powers of Ten? No, he just had that imagination himself right when he eventually saw the movie he realized that other people also had similar thinking um it it's interesting that you know when we talk about complex system science the most really one of the most fundamental and central ideas of complex system science is to be able to think in scale to think in the in the perspective that you look at something

from farther and farther or closer and closer And to be aware of how the different pictures that we see are related to each other. So somehow this is a key way of thinking about the world. And for my father as a child, it was an important part of how he coped with the world at that time. But it's something that has both inspirational significance as well as scientific do you remember the moment you first became interested in a career in science so because my father was a physicist i wanted to be a physicist before i can remember so there really wasn't a time that was that I remember that I didn't want to become a physicist. You received your Bachelor of Science and PhD in Physics from MIT in 1978 and 1984,

respectively. What do you describe as the highlight, intellectually or personally of your time at MIT? I suppose there are a couple of stories. One of them is I came to MIT. So I came to MIT, I was in Israel for a year as an undergraduate, and then I came to MIT afterwards. And when I came to MIT, I walked in and I saw these huge columns, you know, Roman architecture. And I said, this is silly. Why is this? It's not important like that, the architecture.

I didn't come here for this. But the reason why I went to MIT and I enjoyed MIT was because of the excitement of the people who I was studying with, the students and what they were doing. And that's really what made it an important experience for me. There's another experience which is a little bit longer to tell if you have the patience. Howard Bauchner Please. Howard Bauchner So since I told you that I wanted to be a physicist from an early age, there is, when I was in middle school was the time that the tragedies of Biafra happened, Biafra being part of Ethiopia, there was a secession movement. And at the same time, there was a famine. And there were many people who were dying

from hunger. And, you know, and the pictures from that time were coming back to the United States where I was living. And it affected me very badly. And at that time, I spent a couple of weeks really thinking seriously about whether I would continue in a trajectory to work on physics or whether I would switch to working on global hunger. And at the end of that time, I decided that I would keep going into physics, but that I would choose a more practical side of physics that would maybe help, you know, build technology and contribute to economic development and therefore to the situation in the world. So I was, I don't know, seven years old at the time. When I went to MIT, I remembered this, and I, when I became a graduate student, I went into condensed matter theory, the theory of materials, and I studied defects in silicon and surfaces of semiconductors

as a way of doing this practical studies. So having achieved that as an objective, I went to a conference once and I sat down after giving a presentation with someone who was at IBM in the chip facilities designing chips, and I showed him my work and I said to him, does this help you? And he said to me, no, we have heuristics, he said. And that word really cut me badly because I had been chosen this career because I wanted to be helpful. And if I wasn't helpful, why was I doing it? So I decided to switch. And that's actually the origins of going into complex system science, though I didn't really understand it at the time. I really basically decided that I would do whatever I wanted to do. I would ask

questions that were of interest to me. So I started working on other topics, polymers and dynamics and understanding brain and mind issues and evolutionary biology, all kinds of topics that seemed fun and interesting to work on. The surprising thing, if you like, was that a few years later in working on complex systems, I studied the dynamics of food prices because of the financial crisis. And I don't know if this is a topic you want to go into later, but we basically showed the reasons that there were high food prices that caused lots of problems around the world. We can go into the details later. But because of that, I ended up at the World Economic Forum and someone noticed that I was coming there and they invited me to the World Food Program's dinner. They have a dinner there every year. And Desmond Tutu was

the honoree. And I was sitting at the table at this dinner and I basically remembered this story. So you have to understand that this was, you know, whatever decades later, right? This was in 2012 if I remember correctly, that I was at the World Economic Forum. So I'd actually forgotten about this story. But sitting there I I remembered. And it was very sort of powerful for me to talk to the people who were actually involved in distributing food around the world. And to tell them that were it not for a decision that I made at age seven, I might have been on the other side of the table working with them. So it was kind of a very powerful experience for me at that moment. In your book, Making Things Work, you write, quote, although Darwin's theory of evolution discusses how the fitter offspring tend to survive, the reality is that whether or not an offspring

will survive is mostly a function of chance due to the many possible wrong choices that exist for each right choice, end quote. Has that insight influenced how you live your life? Wow, that's quite an incredible um, I think there is this interplay that, um, we don't appreciate between always appreciate between, um, the, um, sort of careful, thought out, logical approach to design, engineering and planning and so on. And the effect of a selection and randomness that's an essential part of evolutionary design, right, is how things get designed. And I think understanding that affects how I think about the world and what I understand as being sort of effective approaches to the world. But the question that you're asking, you know, seems to ask, you know, well, if the world is random, then, you know, how should we think about what happens in it? And I do think that that's an important insight. You

know, people, you know, there are, you know, people who win the lottery, and then it's kind of obvious that it's happened by random chance. and that, you know, they didn't merit that particular thing. It's just, you know, how it happened. Um, um, and I think there are people who are successful, you know, maybe they have a lot of money or a lot of recognition and so on. And maybe they are very proud of their accomplishments and think that it's due to them rather than, you know, that there is some random chance that happened to make it happen. I think that there is a modesty that one should come to the world with that suggests that, you know, things are not always due to one's actions, but due to, uh, randomness. Um, I don't know if you want a little bit more about this, but there's a different kind of, uh, discussion that might be, uh, insightful. Yeah. Please elaborate. So, um, so there's a deep topic of thinking about cultural differences

um and cultural differences are manifest at root in the literatures of each country or each culture um and i've thought a little bit about you know how they differ and this is not a domain of you know of strong you, I've not spent years studying this, but I've thought about how, you know, different literatures have kind of different essences. And as a contrast, you know, describing this, the British literature, you know, you think about Great Expect expectations or, or, uh, you know, other books, classic books, um, the success of an individual is based upon chance. Whereas the U S literature is based upon a heroic idea of the individual accomplishing

something. And you can contrast those with Russian literature, whether you're based on suffering and German, you know, opera, maybe, or literature, which is based upon suicide. I mean, each literature has its own narrative of the hero. And I think that that affects how we think about the world and how we engage with it. Okay, you know, time to talk about the science of complexity. To set the stage, at the turn of the millennium, Stephen Hawking was asked the following question in an interview. Quote, some say that while the 20th century was the century of physics, we are now entering the century of biology.

What do you think of this? End quote. Hawking responded to the question by saying, I think the next century will be the century of complexity. Complexity has become somewhat of a buzzword in this day and age, as I'm sure you've noticed. Before we talk about complex systems, let's firstly define complexity. In a 2019 paper titled An Introduction to Complex System Science and Its Applications, you offered an elegant definition of complexity. You wrote that the complexity of a behavior is equal to the length of its description. Can you elaborate on that definition for us? So this goes back to what information is, right? And there's a subtlety, which is that information should be understood not as what's present in a message, but what's present in the set of possibilities that the message can have.

And to understand that, imagine that you have five characters. Each of the characters can have a certain number of possibilities. Let's say 26 letters, and we don't count capitals and lowercase or punctuation. So if you have five characters, you have 26 times 26 times 26 times 26 different possibilities. There's a lot of possibilities. But if you have a longer set of characters, you have many, many more possibilities. Right? So the, the information or really the amount of information that's present in a string of characters is fundamentally linked to the number of characters you have. And this was an idea that was developed by Shannon in his work on information theory in the late 40s, 1940s.

And that idea has tremendous implications. As long as the number of possibilities is the same, you know, we just take a long enough signal so we can send the same number of possibilities, you can communicate the same information. You just translate it from one way of coding the information to another. So we can use bits to write words. We can use bits to write colors, whatever it is, it's all just information. So information is present in the space of possibilities. anything from a simple entity which you can figure out what it is maybe a block sitting on a table to a complex more complex entity like a human being or an even more complex entity like the society in each case the reason why it's complex is because of the number of possibilities that that system can have.

If it only has one possibility, it's just a simple system. If it has many, many, many possibilities, then if I want to tell you about it, I have to send you a very, very long message. So the complexity and information are really kind of the same thing. There are some subtleties, but that's a really good way to start. So the complexity of a system is the length of the description that I need to send you in order to describe it. Great. Moving to complex systems, Herb Simon had his famous definition in his 1962 paper. He said, a complex system is a system made up of a large number of parts that interact in a non-simple way. Can you unpack that definition for us and give us some familiar

examples of complex systems? The problem is that we actually have to link it to the issue of the multi-scale, the zooming in and out that we talked about before. And the reason we have to do that is that if we have a system that has parts, so let's say we have a whole bunch of light switches, or you can think of it as the same as a whole bunch of bits, which in that sense is just like light switches. And now the light switches can be, they're all the same. They're either all up or they're all down. So there are only really two possibilities, even though there are all of these light switches.

Okay? So that's a simple system, even though you have all of the light switches. Mm-hmm. Another possibility is that all of them are independent. So each light switch can be up or down. And that has many, many more possibilities. So you would say that that's a really complex system. But there's a trick that we have to use to think about it, which is that if I walk away from the light switches. but I only see kind of a little bit fuzziness, I only see them in a sort of,

I see them, you know, from a distance, I won't see each one, then actually, if I don't see each one of them, then I don't see anything, because each one of them is independent. I don't know if that's clear. I'll give another example. But the point is that if I don't see any one of them,

that standing from a distance, it again looks very simple. So another example would be imagine that they're interacting with each other so that the one switches that are near each other tend to be in the same orientation. Okay? But if you go further away, then they're in a different place. Now, I have fewer possibilities when I go up close, but I have more possibilities standing from far away because I see the waves and differences as it goes along the wall. The trick about understanding complex systems is that what's important is not just whether there's a long description, but whether there's a long description in scale, at different scales.

And so the importance of the interactions is that it gives rise to this ability to have behaviors at different scales. Let me give another example. Imagine that you take a person, say me, and you put me into a blender. Excuse the analogy. And yes, you push the button, and I end up as this murky liquid, right? So the murky liquid has molecules, atoms actually, atoms that are moving independently.

Because they're moving independently, like the light switches that are all independent, there's many, many possibilities that those light switches, that the molecules or atoms, sorry, can be in. And that's the situation where it has the highest entropy. Entropy is disorder, and entropy is also information. It's the same quantity. It's the amount of information to tell you where each and every one of the atoms is.

Now that seems very high complexity, but it's only that complexity if I look at it very close up, so I can see each atom. But as I move away, what do I see? Hey, it looks like a murky liquid. That's nothing to describe. On the other hand, if you look at me, there are all of these is moving and I can describe what all of the cells are doing and the molecules are doing and the atoms are doing. Is it clear? Okay. On the other hand, looking at a large scale, I can see something. There's stuff going on. So the trick is that a complex system is a system that has interactions that create these collective

behaviors that you can see at all scales or at many scales okay so it's the the the idea that herbert simon was talking about about systems with interacting parts that's what makes them interesting yeah if they're completely independent they're boring if they're completely dependent like all of the light switches either up or down that's boring but if you have interactions that give rise to all of these different kinds of collective behaviors that's complexity got it so another way to describe those different options is in the language of the mathematician Warren Weaver, who distinguished between problems of simplicity,

problems of disorganized complexity, and problems of organized complexity. And one end of the spectrum, we have organized complexity. For the purposes of the spectrum, we have organized complexity. For the purposes of this discussion, we're not interested in that. On the other hand, we have problems of simplicity. For the purposes of this discussion, we're not interested in that. But in the middle ground, we have what he called organized complexity. And that's where at different scales, you can see complexity. Exactly right. Okay. Now, we should introduce another piece of jargon,

which is the word emergence. And I want to quote from a great 1972 article by the physicist and yet-to-be Nobel laureate Phil Anderson, published in Science. The article was called More is Different. And in that article, he said, quote, psychology is not applied biology, nor is biology applied chemistry, end quote. What was Phil talking about there so what he is really referring to is the way i would explain it is that physicists and you know biologists and chemists spend many years taking things apart

and looking at the smaller and smaller components and the idea of that was that by looking at the smaller and smaller components we're looking at the fundamental nature of the reality around us mm-hmm right you see of this complicated stuff now if you take it apart you look at the pieces and then you take the pieces and now we've figured out the pieces. We have the elementary forces. So we know how everything works. And what he was saying was, um, that thinking about how things go back together is a different endeavor. It's not the same endeavor. And amusingly enough, before he wrote that article, actually, when I did my thesis, which was before I was a complex system scientist

altogether, but I was studying defects and surfaces of materials, I said something not quite what Phil said, but something similar. I said that we shouldn't think about the deconstruction of things as the only fundamental inquiry. That there is a fundamental inquiry in understanding how things come together. And what are the principles and the essential ideas that are needed in order to understand that. And so what Phil was talking about is this idea of emergence where, right, you have all these pieces, right? We have all these atoms or these quarks or whatever we're talking about. And we put them together and all of a sudden we have psychology or we have biology. And you cannot talk about biology and psychology in terms of an elementary particle.

You have to understand how they come together to create behaviors that are very, very different in many ways from the behaviors of the elementary particles. And that's what emergence is all about. Got it. So he was attacking the constructionist hypothesis, which was the notion that assuming we succeeded in reducing the universe to its simplest fundamental laws, we could then reconstruct it from those laws but he said that was a fallacy because of this idea of emergence which is to say the important information in a complex system resides within the relationships between the science of understanding how the relationships work right what what they mean and how they impact on the behavior of the system is really what complex system science is about and going back to what we talk about the light switches it has to do with

this space of possibilities and complexity so the all of these ideas are linked to each other hmm so he ends the article with a couple of economic anecdotes which which I like maybe I'll just read them out to help make this concept even more clear for everybody right after all the title of the article was more is different the first anecdote is Marx when Marx said that quantitative differences become qualitative ones. And then he ends with the famous dialogue from Paris in the 1920s between Fitzgerald and Hemingway, where Fitzgerald says, the rich are different from us. And Hemingway replies, yes, they have more money. Yeah. So I think this idea of qualitative difference is actually structurally important. And the reason is that we have this intuitive understanding,

and a lot of our intuitive understandings have been set aside because of traditional mathematics calculus and statistics which which complex systems is really about going beyond those but you know we have this intuitive understanding that a mountain and a molehill are different. Right. So that's why we use that phrase. Right. And so the point is that there are differences that matter in a fundamental way. And understanding which differences matter in a fundamental way is part of what enables us to understand the world around us. And a lot of that has been set aside in the, in the effort to quantify, um, the world around us using traditional calculus and statistics. And one of the, you know, because the world has dependencies and because it is multi-scale, because there are all these different behaviors, our brains are actually designed for the world around us to understand the a lot of our scientific thinking to like a two-dimensional part of what is really a three-dimensional space.

So one of the exciting things about doing complex system science is you step out of this two-dimensional confinement and all of a sudden you can think about all of these new things that you couldn't think about before and that's both exhilarating and incredible but you know one of the narratives about this is that if you tried to do this early on in complex system science other people couldn't understand what you were saying and that's what I experienced so it was very hard to communicate right going back to this problem of communication it was incredibly difficult to translate the ideas that complex system science generates into a language that could be understood by those who were confined to this two-dimensional

thinking of traditional math. Yeah. While we're talking about emergence, do you want to mention phase transitions? Absolutely. So in physics, the breakdown and the reason why calculus and statistics were understood not to work came because of phase transition. So the easiest example is water boiling, right, because we've all seen water boil. So a phase transition is you sort of change some parameter, like the temperature, and then at some point the system changes radically.

It goes from one kind of state to another kind of state. So it's a really exciting thing. So physicists studied this. And it turns out that, and people may know this, that if you raise the pressure enough, water doesn't boil. You can raise the temperature and you go from something that's water-like to something that's gas-like without ever undergoing a phase transition. The reason is that as you increase the pressure, you squeeze the molecules together so much that there's no difference between rolling around and bouncing around.

So if you follow this transition of water to vapor and you go up and up in pressure, at some point the transition ends. And that point was called a second order phase transition point. And physicists studied this because it kind of has interesting properties. And if you walk along the line of the transition, there is this discontinuity in the density.

So how much is the difference between the density of water and the density of the vapor? And that discontinuity has to go to zero at the second order phase transition point because it ends. So if you look at how that behaves, it's zero and then it grows. There's a certain curve that happens that turns out to be a power law. Okay. So it has a form X to a power and the power can be derived by a theory that was developed by landau russian physicist and he can derive it completely without any assumptions except for calculus and statistics and it's a half so you get a half that's a great answer the only problem is that the experiment gives you something close to a third which is pretty pretty different to a half. Which is, yes, it's not the same, right? And if you get a different number, you're in, you know, something is wrong, right?

You can't just say, hey, you know, it's okay. It was kind of almost right, right? Close enough. So physicists were really troubled by this. And the solution to the puzzle was discovered by Ken Wilson in 1970. Renormalization theory? It's renormalization. Right. What do you want to say about that? So the point is that what's happening at that second order phase transition point is that

it's not, the problem is in some sense not the equation. It's what variables you're using to describe the system. What's happening at that point is that there are fluctuations. It's water-like and vapor-like. And those fluctuations are not just microscopic. They go all the way to macroscopic scale. You can see them in light scattering. It becomes a little bit opaque. So what happens at that point is that you can't actually describe it just with a density. And because you can't describe it as a density,

you need a different math. Now notice that we've incorporated a number of the different ideas we were talking about. Emergence, right, because we have this collective fluctuations. And the interactions between different parts that give rise to these fluctuations. And so all of these are really present in this phase transition context

that was discovered in physics in what seems to be a simple system, which is just water. Okay. But basically what was reason that it happens there is that it's an order to disorder transition. The molecules in some sense are ordered in one place and disordered in the other place. And every complex system has this behavior of going from something where it's more order to less order, or that the order or the dynamics of the order matters. That's the behavior of emergence. Got it. So the mathematics of renormalization group is a tremendously powerful tool for thinking about complex system science and all of these issues that we've been talking about. What has to be generalized to a multi-scale view, right? This zooming in and zooming out,

but the core ideas of renormalization group are the ones that enable you to deal with this circumstance in a powerful way. The second key property of complex systems we should talk about besides emergence is self-organization. What's self-organization? Well, you want a whole course on complex systems. I said this was going to be a crash course. We're doing the whole thing. Oh strap in people yeah so self-organization yes what should we know about that so um the really, um, key idea here is, you know, if you see something, um, you see a painting, it's a beautiful painting. Um, how did it get there? And it got there by someone putting dabs of paint in an intentional way, Unless, you know, it's someone

throwing paint at it, right? I mean, there are other ways, but it's an intentional process of putting things in a particular place. And that's also true when we do engineering, right? We put these parts and we put them exactly where we want them to be. Or when we build a building, we put things in place and so it makes sense that a building has structure because someone made the structure happen but let's say you look at a um a ripple in water or clouds in the sky or um um well, I mean, if you look at an animal or a plant, the question is who put the pieces in exactly the right places in order to achieve the structure that you see?

And the answer is, nobody did. There was no intentional act where someone put all the molecules in the right places. There's this weird thing that we say, which is, you know, we talk about, you know, the structure of biology is coming from the genome. But we say that there is a blueprint. We talk about the DNA blueprint. So if we open up the cell, what we're supposed to see is that there's this little diagram with arrows and things to tell you how to put the structure together. It doesn't work that way. So how does a single cell that is structuralist to look at it become a human being after a process of development in the womb? Who put the pieces in the right places? And the answer is, what really happened

is that the cells talk to each other, right? So, you have one cell that divides or multiplies. In biology, it's the same thing, right? So you have these many, many cells that form, and then they talk to each other. And that conversation happens that they decide in some very simple ways, according to simple rules, that one is going to be a brain cell or a heart cell or a or a heart or also muscle. But it could be a nerve cell or it could be a tooth, you know, part of a tooth or all kinds of things. Right. The question is not that they decide to do that, but they also decide where to be. That's an incredible process. So that's the process of self-organization. The self-organization happens because it arises because of the interactions of the parts. And so now you see how that's related

also to the emergence idea, right? Because interactions give rise to these multi-scale structures of these patterns, right? So patterns is another complex systems word. The funny thing about this is that when you do statistics they tell you that patterns are deceptive they're there in your mind because statistics doesn't describe patterns so they dismiss them but it's obvious that they're not just in your mind right people are not just random arrangements of stuff so patterns that exist in the world arise because of interaction. And that's a hugely

powerful idea. And it seems all mysterious. And in fact, it was so mysterious that for many years, people believe in what's called the homunculus theory, that there was a little person in the cell that grew and grew and grew and grew until it became big. But of course that's not the case. It has to do with the emergent behaviors that arise because of interactions among molecules and cells to create elaborate, intricate structures. Got it. So emergence we can think of as occurring across scales, whereas self-organization we can think of as occurring over time. Yes. Perfect. systems whether the system is consciousness or the stock market or the weather there are things we can distill that are common across all of their

systems yes absolutely and and and and this universality that you're talking about has many different itself meanings right because one of the ideas of universality is that the same principles or ideas apply across all of these systems, physical, biological, social, right? There's self-organization of weather patterns, creating hurricanes. And there's self-organization of biology, creating patterns on animal skins as well as the structure of animals. And there's dynamics of stock markets and fads, you know, people buying skinny jeans or something like that. more powerful form of universality where you can construct key mathematical representations representations of real world systems that apply across different contexts i find that kind of weird in and of itself the principle of universality

does that strike you as as spooky or interesting so it's it's kind of built in i mean it's actually it's actually you know the idea of universality was sort of elaborate on and had a formal meaning when renormalization group right dealing with phase transitions came up but it's actually much more fundamental than that so for example for example, in basic physics, we learn about pendulum. Yeah. Right. Or an oscillator. And we learned that the same math describes music, right? The vibrations of strings. So why does the same thing describe that and also describes cycles, right? You can use it to describe, you know, human cycles or, you know, I mean, in a simple way, you can use it to describe the daily cycle, right? Everything is cyclical, also is like an oscillator in some sense there's cyclical behavior

so we use the same math to describe these many different things so that's really what universality is about now why does it happen and the reason it happens goes back to this multi-scale idea so again we're connecting these different kinds of ideas If we go to the largest scale behavior of a system, so the thing that you can see when you're the farthest away, but you can still see something, right? Then you can't see a lot, right? You can only see a little bit of what's going on. But if you can only see a little bit of what's going on, it can't be very complex. So then it can only be described by a few things, mathematically a few things, like one variable, two variables, three variables. Because it can only be described by a few variables, you can't have many mathematical equations.

There's only a few mathematical equations with a few variables. So if you only have a few equations, then every system that exists in any context has to fit within one of those classes of behavior. So that's why they're called universality classes. It's like these pigeon holes that you can put every different system in, and they have to have the same equations because there are only a few equations so you're done okay yep that makes more sense to me now before discussing a few applications of complex system science i want to by the way i have to say i'm impressed that you've done all of this in in the in this time it's not easy to

think about all these ideas and i'm sure that it's exhausting for people thanks yeah yeah well hopefully um i'm sure everyone will will uh be be patient with us but the the crash course is is reaching its conclusion uh and it's And it's been absolutely fascinating so far. But I do want to drive home this point of scaling because I think it's going to be so important later when we talk about the spread of the coronavirus. Are there any great examples of why scaling is important that you can offer just to make the concept more concrete?

I've got here Jeffrey West's book, Scale. He talks about whether it would be possible to have a giant human like you see in the movies. Is there anything like that that can help people intuitively grasp the importance of scale? Sure. So one of the important ideas in complex systems is that there are some systems that are the same in some sense at different scales. So for example, a classic way of thinking about it is you look at a tree, right? A tree has a branch, the trunk, and then it has a branching. it is you look at a tree, right? A tree has a branch. The trunk, and then it has a branching. And then you look at the smaller branches, and if you zoom into them,

they look kind of like the bigger branching. And then you look in smaller, and you see the twigs also have the same kind of branching. Also, you have, you know, sort of blood vessels in your body. You know, you have big blood vessels and smaller blood vessels and smaller blood vessels. Another example of this is Mandelbrot, who you haven't mentioned, right, the coastline of England, right? There's, if you look, it's rough, and then you look closer and it's still rough, and it's closer and still rough. And that violates, by the way, one of the assumptions of calculus, right? Because usually in calculus you assume things are smooth, you go fine enough and things are smooth. So if you have many, many scales at which it behaves in a rough way, kind of calculus no longer is a good tool for thinking about the system.

Now, so that's one way in which we have multi-scale things. And you can think about in the market, right? So the market, it has fluctuations, right? So if you look at a minute, there are fluctuations. Or at a second, there are fluctuations. Or if you look at a day, there are fluctuations. Or a year, there are fluctuations. So no matter what scale, it doesn't become smooth. So that idea of sort of the non-smoothness of systems is really crucial for thinking about systems. But the multi-scale idea generalizes that because sometimes it does become smooth

right so you you know you look at the earth and it's rough and it's rough if you respond but as you get eventually you just have the earth and it becomes a point if you go zoom in to the blood vessels, eventually you have the smallest possible blood vessels because beyond that you have cells. So there are some things that have a natural size, like people have a natural size. And there are some things that don't have a natural size. And then you have many different scales at which those things happen. And if you combine those two ideas, you realize that you have to describe things in a multi-scale way. So that's kind of a very sophisticated understanding of what we mean by multi-scale thinking.

Got it. So on the one hand, we have fractals like coastlines or capillaries. And then on the other hand, we have multi-scale emergence where whether we zoom in or out, we start to see different levels of information. Correct. Okay. Wrapping up this incredible crash course, let's talk about an application of complex system science because you've been incredibly active in solving real world problems using the insights of complex systems.

On this podcast, many times before, we've spoken about speculative bubbles and informational cascades. So maybe I'll ask you about a different example. We could talk about the Arab Spring, or we could talk about the high food prices, which you alluded to earlier? So in 2007, when, um, so the world, um, went through a crisis, financial crisis, um, uh, we had these tools of complexity science and we went to investigate what happened. And one of the things that we wanted to understand was kind of the causes and consequences of the financial crisis. And, you know, there are various different ways to tease this out. But one of the things that happened a few years later was the Arab Spring was this incredible, you know, seemingly spontaneous appearance of riots and revolutions in a substantial part of the world across multiple countries. And if you ask people, you know, why it happened, most people would say that it had to do with sort of the governance, the dictators, you know, the dictators that were there. But those dictators were around for a long time. So it's really hard

to pin it on them, if you will, in terms of the dynamics. And what we found when we looked at the dynamics of what was going on is that food prices were the trigger. And in fact, when we did the investigation, what we saw was that food prices increased and we looked for the consequences of food prices. And what happened is that food prices increased dramatically. They doubled in 2007 and 8. This is just after the financial crisis. And then they went down, and then they went up again in 2010 and 11. And just before the Arab Spring happened, we sent a warning to the government

about high food prices causing social unrest and political instability. And basically that was a prediction of the Arab Spring four days before Mohammed Bazezi started things in Tunisia. So the rising food prices triggered the Arab Spring. And so how do we understand that? Where do we where do we anchor the behavior of the system that created that rise in food prices? And so and again, rising food prices causing food riots is not surprising. Causing revolutions is not surprising. But the key thing is, how do we understand where did the food prices come from? And it turns out that there

were a lot of things that people talked about, you know, and, you know, relevant to your audience, there was a drought in Australia. At least people claimed there was a drought in Australia. It turns out that if you look at the data, there probably wasn't really a significant drought. But the idea was that it reduced the production of grain and that caused an increase in food prices. So it has to do with the supply and demand. But it turns out that if you look at Australia and the global system, and I don't want to put down Australia's agricultural production, but it's not big enough to cause this kind

of big effect. So if you did a statistical analysis, you would put in Australian food price, Australian grain production and 25 other factors and you do correlations and figure out, you know, what was happening to cause food prices to increase. But if you have a fly and you measure the height of a fly over the ground and you correlate that with food prices you might get a correlation but there's no way the fly is causing the food prices. So conventional statistics don't work when we're analyzing organized complexity. Right. And it turns out that there are only two things that cause the food

prices to go up. One of them was a supply and demand factor, and that is the ethanol rules in the United States that caused a massive amount of corn to be diverted from food into gasoline. And that's a big effect. It turns out that it caused a doubling of food prices from when they were instituted in the early noughts, 2005 and 2007 until like 2012, 2011 or something like that, they caused the doubling of food prices. It started a little bit earlier and then it went there.

So over a period of less than 10 years. But the other thing that happened was crashes, market bubbles and crashes. And the reason for the bubbles and crashes. And the reason for the bubbles and crashes that they were so large is that commodities were deregulated in 2000. And if you care about politics, the rules on ethanol were instituted by Bush Jr., and the deregulation of the markets was a Clinton mistake. And he admitted it when I spoke with him later. But when you deregulate markets,

what happened is that it enabled many, many people who wouldn't otherwise be able to put a lot of money into commodities. So when the markets crashed, right, the first mortgage market crash and then the stock market crash in 2007 and 2008, people needed some place to put their money. And there are only really four markets. In addition to mortgages and stocks, there's also bonds and commodities. So a lot of money flowed into bonds and into commodities. And commodities are a tiny market.

They're a few hundred billion dollars. And a huge amount of money came from trillions, tens of trillions of dollar market. And basically the flow of the money, which is not a supply and demand effect, but is a dynamical effect, caused the prices to create a huge peak. And that causes food riots in 2007 and 8. And then there was an oscillation. Remember, we talked about oscillations. There was an oscillation here that led to a second peak in 2010, 11, right? There was a crash and then another bubble. And that second peak is what triggered the Arab Spring. So here's the kind of pattern of this global crisis. We have these change of rules in

the United States, plus a mortgage crash that creates this flow of money into the commodity markets. And that creates high food prices, which triggers the Arab Spring, which leads to the disruption of governance in Syria and a couple of other places, which leads to the nucleation of ISIS, right? That was where it was incubated, which led to the European refugee crisis and all kinds of other things that are happening around the world. It's a pretty incredible cascade from economics to social context to people not being able to buy food. Now, why were we able to model this?

What did complex system science bring to this analysis? Surely it wasn't the recognition that high food prices cause riots. So one of the things, the really crucial thing, is knowing how to identify what are the most important few variables and the right equations for those few variables. This was this universality business that we talked about. So because we were able to do that, we can describe dynamics. We don't have to describe it with a thousand variables. We can describe it with a few.

Turns out there were four variables in the equation. Wow. And we could, and it was a deterministic model that we used to model this. And the modeling of food prices had a statistical validation of 10 to the minus 60. Right? That's like unbelievably winning some incredible lottery, right? So we really understand what's happening because we first wrote down the equation and then we fit it to the data it's not like we we tried many different equations it was only one equation that we tried because we knew what the right universality class was and so we wrote down the right equation

and it fit perfectly That's incredible. Coronavirus. Let's do it. Let's not do it, hopefully, eventually. Yeah, yeah. Let's talk about it but not catch it. That's right. So as of recording, it's officially named SARS-CoV-2,

but maybe we'll just refer to it as coronavirus for simplicity. And the disease that it causes is named COVID-19. I thought we should begin by taking a quick inventory of what we know so far. So I checked the Johns Hopkins University statistics this morning at 9am on Sunday the 8th of March, Sydney time, and we had 105,820 average an infected person will transmit the infection to, assuming the population has no immunity. What do you think our best estimate at the moment for the R0 of the coronavirus is? So it's about three to four. So for each person that's infected, they infect about three or four other people. And the number is not a very solid number, not because people haven't tried to extract it,

but because there's actually a fundamental reason why it's hard to pin down, which I might explain if people have patients. But that's about the number. But what it means is that it spreads incredibly rapidly. It grows really fast. To explain it, I actually prefer a different number. And even that number is, you know, it depends upon how you think about what the pandemic is about, what the epidemic or pandemic is. The terminology is not that essential here. The key thing is how rapidly does it grow in time is one of the key things, as well as how many cases, right, this reflects how many cases. And I prefer to think about it in terms of not the number of cases that one person infects, but the number of, the change in the number of cases from one day to the next. Why do you prefer that?

Because it's easier to think about the dynamics. If you want to know what's going to happen a week from now, it's easier to think about that than to think about the number of people because the number of people one person infects. Because you have to go from the number of people one person infects to how much time does it take for them to infect them, you know, and so on. But if you want to think about sort of how rapidly it's growing, you know, how fast things are happening, then it's easier to think of one day to the next. Okay. And from one day to the next in at the time when, you know, so it started, of course, in China. And on January 23rd, China imposed the lockdown on Wuhan and neighboring areas. And just before that and just after that, because they didn't you don't stop anything when you act immediately because they're all the people that have already become infected that haven't been seen yet because they're still without symptoms. During that period of time, the rate of change from one day to the next was a factor of 1.5. That means that if you had 100 cases in one day, the next day you had 150.

And the next day you had- Howard Bauchner 225. Howard Bauchner 225, right. So every day you get a multiplier of 1.5. 225, right. So every day you get a multiplier of 1.5. And that means over a week you get a multiplier of about 20. So after two weeks you have a multiplier of 400. And if you follow that through, after two months you've infected the world. Just two months.

So it's really fast. That's the way this virus works. There's a very important point I'd like you to elaborate on. In a note with Nassim Taleb and Joe Norman, titled Systemic Risk of Pandemic via Novel Pathogens Coronavirus, which I think you published in late January. You wrote that estimates of the R0 are biased downwards because of fat-tailedness, which has to do with super spreaders. Do you want to just explain how that works? Sure. So the easiest way to think about it is that, you know, some people, most of the time they stay at home and other people go and party all the time. Right. So imagine that you have someone who has is infected. And if they go home, they might not infect anybody. But if they go party, how many people are they going to infect? So there was a

case where someone went to a religious service in South Korea and there are like 3000 people that resulted from that infection. That wasn't the first round, but there are probably tens or hundreds of people that were infected just from that one person. And, um, and and and so the point about this is that there's a there's a significant chance of a large event so when we normally when we think about normal distributions or when we think about sort of averages right this is it goes back to statistics. If you have an average, the chances that you're very far away from the average are small, really tiny. But in this case, while it's smaller, the probability is smaller to get such a large event, it's large enough that it matters. And so imagine that you're trying to figure out what your are not is. So you, you,

you take, you know, you have a bunch of people, say a hundred people and you calculate what the are not is. Well, either it does or it doesn't have a super spreader event. And if it doesn't have the super spreader event, then surely the average is going to be much less than if it does have a super spreader event. So it turns out that even figuring out what are not is very hard because you know it depends upon whether you have you know one or other you know large events and so it's just an uncertain number yeah and i think um during during the sars outbreak there was a famous super spreader event that occurred in an apartment block. Yes.

Was in China or Hong Kong? I don't remember. You may be right that it was Hong Kong. But the point is that I looked at analyses of that outbreak. Yeah. And they described clearly super spreader events, which is an important way of thinking about it in epidemiology using standard statistics. But the other way to describe them is using a power law or fat tail distribution, which has a continuous distribution that includes these super spreader events. Yeah. So in other words, we could wake up tomorrow and find out that the

R nought is actually much higher because one of these fat tailed events eventuates. Yeah. The other way to think about it is that, um, you know, we can now maybe take a quick segue into action. It kind of makes sense. No, it kind sense, it really makes sense to stop large events. Right. So, um, so, uh, um, comic con in Seattle was just canceled and South by Southwest was just canceled. Um, but people have not yet clamped down on theaters and concerts and, and all kinds of, and religious services and so on. And they really should. I mean, why would we want to have a super spreader events if we're, if we're trying to, to stop the outbreak and even to slow the outbreak, but I really want to talk about stopping it. Um, uh, the, you know, cutting off the tail of that distribution really matters tremendously.

You do not want to have an event of hundreds of cases. And the same thing goes for, you know, various ways of interacting with lots of people. You just want to make sure that that's not happening. Yeah, this is crucial. We'll come back to action. We will come back to action. We will come back to that. So to continue taking stock of the virus, you mentioned that it spreads incredibly rapidly. We have over 1.5x new cases per day. What about the fatality rate? So, I don't know how to say this.

I do not understand how people talk about the fatality rate of this disease as being, you know, some people say it's not large. It's huge. There are somewhere between three and four percent of people with the disease that die. Now, this is on the latest China World Health Organization data. Yeah. And again, it varies because, you know, sometimes you get a cluster of people that are infected that are vulnerable and they die. And so and you're not checking for you're not yet testing for other cases. So you have a very high fatality rate. Other times you have a lower

fatality rate because, you know, you did the measurement early on in the outbreak. So you have a very high fatality rate other times you have a lower fatality rate because you know You you did the measurement early on in the outbreak So you've just done a week and of course the people haven't died yet because the disease hasn't run its course So you have some places where you have a low fatality rate someplace you have a high fatality rate and people all say, you know Hey, it's probably not so bad because over here we had this low fatality rate or you know I mean, oh and then they dismiss this other place because you know, hey, that's whatever there's all kinds of arguments But really the solid number which is an incredible number is three to four percent who have died and it's it's and it's not over yet again because of the numbers that of

People that haven't yet been cured in China, right? So it may still go up. If you look at the current trend in the cases, the fatality rate is still going up. So it's probably very, I mean, it's anyway very, it's 3.7%. So basically we can take as a number 4% that are going to die, which is 1 in 25 people who catch the disease. Yeah.

You mentioned that we need to build in lags to calculating the fatality rate. So to explain that to the audience, because the denominator is growing exponentially, you need to, for want of a better way of phrasing it, allow people time to die. So you need to build in some sort of lag or otherwise you're going to underestimate the fatality rate. Does that 3% to 4% statistic build in any lags? So that's the best number that we have from China, but also from other places where it's being done carefully, right? There are a lot of deaths, but the early on, you know, we have to look back in time in order to figure out what's happening. And that's the best number that's available. Okay. Now there is a silver lining in here and that's the Korean

data. So Korea doesn't have a denominator problem. They've been incredibly rigorous with their testing. On the latest data, 140,000 people have been tested. They found 6,000 cases of coronavirus and that's led to a 0.6% death rate. Yeah. So that death rate is, again, most of their cases are in the last week. Okay. So I definitely expect that death rate to go up. Right. But let's say even it's more closer to 1% there. And again, I don't think that that's going to be the case because of the fact that most of the cases are just now. It's still a pretty high death rate.

I mean, it's still 10 times what you get from flu. Yeah. 1%. Imagine 1% of the people dying in the world yeah it's crazy um yeah so that's not the whole story i mean people talk about the death rate and they think you know hey it's only you know it's like you know so i have one percent chance of dying um so the other piece of this is that 20% of people are put into the hospital and are in the hospital for extended period of time. And about 6% or something like that, maybe to 10%, are actually in ICUs. So they're intensive care units in life-threatening situations. So the severe disease, I mean, it's a really severe disease.

So some people get away with a light case, but a lot of people are in the hospital for an extended time. The 10% for ICU admission, that comes from Italy, right? No, it's China mostly. Okay, right. Italy is a different story because they're not really testing, so only the severe cases. So the fatality rate in Italy is actually much, much higher because they're not really testing, so only the severe cases.

So the fatality rate in Italy is actually much, much higher because they're not testing for the cases. Yeah. Now, Spanish influenza came in three waves starting in March 1918, lasting through the U.S. summer of 1919. What do you think is the likelihood that the coronavirus abates seasonality? In other words, how likely are we to see further outbreaks in, for example, late 2020, early 2021? I think the answer is we don't have time. Yeah. I mean, it's so fast. Within two months, the world will be overrun unless we stop it. So, I mean, basically it's I guess do or die is the right statement right I mean if we don't if we at the rate at which it's growing

now if people ranging from individuals all the way up to the government don't take action within two months it may have run its course with 1% of the world dead. And, you know, there may be other things, right? People, there's evidence that immunity is not really conferred by the disease. So people may get it again. So there's a lot of issues with this that are really terrible. But the point is that seasonality, which people talk about and think about, hey, maybe it'll be less when the weather warms up or something like that. All of that is not that important. It spreads rapidly. There's no evidence that it's going to stop spreading when it gets warm, really. And, you know, even if it slows down a little bit, it spreads so rapidly that it doesn't really matter that much.

One approach to thinking about the likely spread of this virus as well as its consequences is to look to historical base rates. For example, what happened during the Black Death, what happened during the Spanish Influenza, and use that as kind of like a benchmark to predict the path of coronavirus. What are the problems with the base rate approach? So I don't know that any of the prior cases are sufficiently similar to this one. And the main reason is just global connectivity i mean um the dynamics are just so fast yeah let me let me just ask you a question about that so a couple of quotes neil ferguson book, The Square and the Tower, quote, The basic reproduction number of a pandemic is determined as much by the structure of the network it infects as by the innate infectiousness of the disease, end quote secondly in a 2006 paper titled long-range interactions and evolutionary

stability in a predator prey system and again in a 2019 article titled transition to extinction you wrote about the risks of increased global transportation what are those risks? So this is what started us working on pandemics. We were studying the effect of long-range connections. So imagine the direct flights that are happening now from Sydney to New York, right? So as the world becomes more connected, we can think about what happens with disease, with pathogens. And you can make a very simple model. Remember, if you're describing the right variables,

then the simple math is what you need. So going to universality, this is what we were trying to understand. What happens with long range transportation? Excuse me. Um, the, the basic idea is what's the difference? What changes between not having the connections and having the connection. And when you don't have the connections, diseases that are very aggressive, they're going to spread rapidly, but if they kill their hosts they're gonna die out because they're gonna run out of the people that are available to them locally or whatever is the thing that they're consuming it's like a predator prey or post pathogen it's kind of again it's universal dynamics it's the same kind of thing So if they spread fast and they kill their host,

they're going to die out and they won't survive. But if you have long-range transportation, then they don't die out, they go someplace else. So what happens as you add more and more long-range transportation is that the system goes through a phase transition, like the boiling of water, to a state of vulnerability to extinction. So it's not just that the pathogen spreads more rapidly, but that the pathogens that are around are more deadly. Those are the ones that continue to exist in a globally connected world.

So we realize this and we basically, I've been talking about this since 2006, and warning about Ebola outbreaks and SARS type outbreaks. Well I believe you spoke to the World Health Organization as early as 2004 about that, right? No, it was actually 2014, to be fair. 2014? Yeah.

I hope I didn't have a typo in one of my articles, which might be. But in 2014, in January, I indeed, I went to Geneva and spoke at the World Health Organization, gave a lecture to a bunch of people there. And when I gave this part of the presentation, it's only about a five minute part of a larger presentation. You could see the jaws dropping in the room, right? I mean, what I basically told them is that their expectations about the future were incorrect. People who think statistically, right, they think that the past experience, right, the accumulated statistics of the events that they know are predictive of the future events that they're going to experience. So Ebola outbreaks

that are small are going to predict future outbreaks being small. And SARS outbreaks that happen, well, that's what's going to happen next. Or the flu SARS outbreaks that happen well that's what's going to happen next or the flu is that's the same thing as what's going to happen next but that's not what happens as you increase connectivity because you're going to this transition yeah and as you go to this transition do you get these big fluctuations remember just like in in in water and these big fluctuations in this case are massive extinction events. So you end up with these massive extinction events. And what I had wanted to do at the time was to really study it, figure out how close we are, prepare for it, do all kinds of things that

would prevent it, et cetera. But nobody was listening, really, which was amazing to me. And at the World Health Organization, while their jaws dropped, and they acknowledged our work, it's not that they didn't respect it, they went on about their business, and Ebola started later that year and ended up being this incredible outbreak that they really didn't know how to stop. It went out of control. And, of course, SARS was another example of a disease that was very powerful. But the coronavirus that we're talking about now, well, this is really, you know, the next level of risk because of the combination of how deadly it is, how rapidly it transmits, the period at which you don't know the symptoms, the fact that it may be infective just before it has symptoms, the specific properties of the disease make it very difficult to stop.

And that's what we're talking about. The vulnerability that we have to the extinction arises because of global transportation. Because the transportation routes enable more aggressive pathogens to survive and spread. That's right. And spread rapidly. Yeah. And that means that in a world of greater connectivity, as you say, as you tried to explain to who, past performance is no guide to the future exactly and so taking you know this is really the problem you know people try to interpret events in terms of the

past and the power of the math that one can use is to generalize right science why is science helpful if you've seen something before you can can do it again. You don't need science. The power of science is that you can generalize. So you can think about, you know, the classic, you know, something falling on earth like an apple in the same way as you can understand the moon. And the fact that we can understand these different things enables us to deal with things that we haven't experienced before in a direct way. And so the ability to understand the coronavirus outbreak by understanding the math is hugely

powerful and very different. And going back to this discussion of the basic statistics, you don't have to understand a lot of math in order to understand what's happening. You don't have to think back. In fact, you want to throw away the old experience and just say, look, you know, this thing is growing and it's multiplying by a factor of 10, 20 actually every week. So what's going to happen next week is really different from what's happening now. A factor of 10 is huge, right? You go from five cases to 50 cases to 500 cases to 5,000 cases. So the point is that you have to anticipate next week as being different from this week. And that's the main thing that you have to know.

Yenir, quoting you from your 2004 book, Making Things Work. Quote, when we think about emergence, we should, in our mind's eye, be moving between different perspectives. We see the trees and the forest at the same time. We see the way the trees and the forest are related to each other. To see in both these views, we have to be able to see details, but also ignore details. The trick is to know which of the many details we observe in the trees are important to the overall behavior of the forest. End quote. What are the most important details for us to focus on in the spread of coronavirus and which details should we ignore? Great. So indeed, there's a lot of argument about the death rate and a lot of argument about the spreading rate and all of this stuff.

And people have this obsession with trying to figure out what the right number is, and it really doesn't matter. The most important thing about the disease is two factors. One is that it spreads, that the question is whether it spreads at a rate that's greater than one or less than one. That's the most important thing. Either it's growing exponentially or it's decreasing exponentially. Everything else is commentary.

The second thing is that we don't want it. Binary statement. We do not want it to spread around the world because it's deadly, it's terrible disease. The third statement that's the most important statement is that we can take actions that will stop it and those three pieces of information are the only pieces of information that you really need to know so it spreads rapidly means we have to act now not next week but today the fact that it's deadly means we don't want it. So we want to do the action.

The third statement is that we can act. Once we have those three pieces of information, don't pay attention to anything else. Because they distract from the most important task, which is making the decision to act. That's it. Let's talk about action. In your note with Taleb and Norman, you wrote that, quote, standard individual scale policy approaches such as isolation, contact tracing and monitoring are rapidly, that is computationally, overwhelmed in the face of mass infection and thus also cannot be relied upon to stop a pandemic,

end quote. In light of scaling and everything we've learned about complexity in this conversation so far, what approaches at the government level should we be taking? Yeah. So again, the main thing is that if you don't think about the network of connectivity, then you can't stop the transmission. So you have to think about how the network works and deal with that. And the main, most powerful tool, without doubt, is to reduce the system to its components, to separate it. And we do that even in a medical context. We isolate individuals, right? Once someone is sick, you don't send them into a crowd, right? At the individual level, people understand this. But what's needed is to do this in a multiscale way, to separate the world into regions that stop the transmission across the boundaries so that you can deal with them locally.

Save some places from the disease while you're working on it. It's not that you sort of ignore the disease where it is, right? I mean, there's this crazy idea, which is historically, it's actually happened in, in, in Japan with a ship, right? You close off the ship and you let the people alone. And of course, the disease runs rampant. That's not the point. The point is that you go into the space and you address the disease where it is. So instead of thinking about treating individuals, you have to think about treating communities. So if a community is infected, just like when an individual is infected, you isolate the community and then you treat the community. Treating the community means treating the individuals that

are part of the community. And that means you have to go in, provide food, provide services, whatever, but you prevent the transmission. It's the prevention of the transmission that we're after. So one very powerful tool is isolation, multi-scale isolation. We have to separate the world into zones, zones that are infected, zones that are not infected, and use the divide and conquer strategy for the disease. The second most important strategy is time. That means detecting the sickness early so that we can isolate the people before they can infect others. And that means having early symptom tests.

So in West Africa when Ebola ran rampant, the simulations were saying that it was going to go to burnout, which would have meant 10 million people dead in that area. Okay? But it didn't happen. It stopped at a several thousand. It was a slower moving disease, obvious than the coronavirus. And the reason that it stopped, I think it was, I don't remember like 10,000 people were infected, several thousand, you know, something like that. I don't remember exactly. But the reason that it stopped was that people went door to door and detected when people had fever in the community. And actually, the people who did it were members of the community. So they went door to door with forehead, you know, infrared thermometers.

And as soon as someone had fever, they were isolated. They didn't wait to know even if they had the coronavirus, the Ebola. The point is that isolating communities, they're, if you can detect people early, then you can stop the transmission by isolating them. So you can take this strategy, rather than isolating just people that you know have Ebola, you isolate people that have early symptoms, even if they have something else, and then you still stop the disease. Similarly with coronavirus,

if you stop everyone that has any kind of symptoms from being in touch with anybody else, you will probably stop the disease. So early detection, isolation, even if you don't know it's coronavirus, is the way to go. And so one possible solution to the problem is actually to have massive testing, right? Testing, obviously, of anybody who might have related symptoms, not just the ones who are the most likely to have the disease, because what you want to do is stop all of the people that have the disease. And all of the conversations that I've had with people telling people, you know, hey, we really can take action. You know, we can stop the Ebola outbreak by doing community activities, community engagement. China executed on those ideas

in an incredible way. They shut down transportation. They isolated people by a lockdown in the community. And after starting that with only a few hundred cases, they went to thousands of cases because it's spreading so rapidly. But at this point, they're down to under 100 cases in all of China. In fact, today, 24 out of the cases that they have, which is one quarter of the cases, are actually people coming in from outside. And they're not letting them go into the community. They're quarantining them. So they're really not a threat. So the situation in China has become a medically unimportant problem. They're on the way to solving it. They're down to like 75 cases and they're only in Wuhan. So geographically it's shrunk down to one city and they're on a

trajectory to eliminate it. And so they realized that they did the wrong thing early in that they tried to pretend that the disease didn't exist. But once it really took off, they responded aggressively. They responded in a way that was convinced. They were convinced that they needed to take action and they were able to stop it. Now, South Korea has done very similarly. There was a super spreader event and they immediately locked down the city, Daegu, that had the most cases and they took very dramatic action. And now they're in a declining trajectory. The point is that we know not just by my theories, right, but we know by actual action in China and in South Korea,

that we can stop the outbreak. And Singapore is also a good example. They've been able to control it, you know, not by as radical actions, but they only had a few cases. So with a few cases, you can do different things. But the point is that we can stop it. And the ways to stop it are exactly the ways that we're talking about. Number one, shut down the transportation, isolate people that are likely to be sick, do massive testing, prepare the ability to deal with a massive number of cases, and then you can stop it. The sooner you do that because of the dynamics, the better. Because if you wait another week, you have 10 times as many cases. Another week after that, you have 100 times as many cases. So why not do it now and not later?

That's the point. There was a 2007 study of the Spanish flu that examined school closures and other public interventions across 43 major U.S. cities. And essentially it found that the earlier, the better. That was the conclusion, which is obviously not surprising. But interestingly, one of the additional upsides of having kids stay home is that it typically causes their parents to stay home, too. Right. And so that stops the parents spreading the virus to each other by going to work or catching transport or whatever they'd otherwise be doing. Do you think it's likely that Western governments will be able to pursue draconian measures like the ones you've been advocating? Okay, so here's the thing. Um, um, I can only say I wish they would. And at this point I am,

um, you know, I've been fairly hopeful that when the disease really showed up that they would act. You know, there is a small, you know, reasonable probability that it wouldn't be a major outbreak, right? There were three smaller fires from China, right? South Korea, Iran, and Italy. But what ended up happening is that the Italians are not aggressively pursuing cases. The Europeans aren't shutting down borders. The U.S. has restricted travel from Iran because, hey, it's Iran, so why don't we restrict travel? But they haven't restricted travel from Italy or South Korea people just walk on the plane and walk off the plane

and I'm going here like you know hey this is pretty crazy not to mention the the huge bungling on the part of the center for disease control about not making not making testing unavailable. Ben Hunt of Epsilon Theory characterized the policy as don't test, don't tell. Well, that's happening in Italy, it seems. Other countries in Europe are doing more testing, but I honestly, we don't, we were trying to find out about all of what's happening in different countries. But the situation is severe. There is some, you know, one positive thing, I mean, again, the two really positive cases

are China and South Korea. There is some positive in that the growth curve is not exponential in Europe. In the US it is exponential. In Europe it's not, at least not in other European countries. Uh, and, and Italy is kind of not clear right now. Um, and that may be because they're doing more testing. I don't have enough information about it. Um, but the, but they could save everyone, right? Save many, many lives and many sick people being severely sick if they would take more aggressive action. And the thing about it is, you know, if you actually isolated everybody for one incubation period, not for one incubation, for one infection period, right, the disease

would be gone. Right? So in China it happened over a little bit more than a month basically. So it doesn't take that long. So I'm going here like why not do this? It won't take very long. You know, let's take a vacation for a month. We'll make sure everyone has enough food and water and we'll be fine. And then afterwards, we can go on in our lives, our lives, because we're alive and we're not sick in hospitals or dead.

So the opportunity is now. And all we have to do is stop for one month. It may take six weeks now because we didn't do it earlier. So the challenge is to get people to think dynamically. They don't think dynamically. They say there's such a social cost. We'll stop everyone from going to work for a month. Wouldn't that be terrible? And the answer is, what do you mean terrible? If we stop everyone from going

to work for a month, we save, you know, take your pick, a hundred million lives and you know, in a billion people in severe disease. I just don't, I don't get it. There's a piece of this that doesn't make any sense to me. And I think they, they have this base assumption of, of persistence, right? If we cause people not to go to work today, then it'll be the same forever, right? The social cost will be forever. They don't have a dynamic perspective that if it takes a week, let's do it for a week. You know, if a week we can knock down the disease and then, and the real way to do this is to do it geographically, right? So we do it in the places where it's highly infected. I, you know, they're right. And we don't have to do it in other places.

So where there's no cases, we don't do very much. Where there's a few cases, we do contact tracing. Where there's many cases, we do lockdown. And then we're done. Right? So now the next question, which I really am going to have to address is, So they're not doing it. So what's the alternative? And this is where we have to make choices individually. And do you want me to keep going or do you want to say something? No, keep going. So the individual choice is whether individuals are going to take action. And the way things are going, let's portray it very clearly.

One of the few places that's going to survive without the disease is somewhere in the outbacks of Australia. Right. So everyone should go out there and find a square mile of territory and camp out. Okay, there's a lot of territory there, so it's a good thing. And all they have to do is wait for a couple of months. The world will go up in flames, and the disease will be gone, and they'll be able to get back and rebuild society. There are all of these stories of the end of the world and what it looks like and how it looks to have some people that survive. Well, that's what

we're talking about. Now, it doesn't have to be quite that extreme. What we really have to do is think about the creation of safe spaces for people that won't get infected. So today, after watching the governments do nothing, I put out a piece on basically a guide to families and to the creation of safe spaces, because you don't want to really be by yourself. You want to be with your loved ones, right? And so you have to create spaces where people can be safe. And that's the start of creating a geographical system of of safe zones. And so we have this color zone scheme where you have green zones where you're safe and then yellow zones where you have to be super careful. And then there are orange zones where you have to not only be careful, you

have to do stuff that's aggressive. And then red zones, we have to do lockdown. So the idea is to mark a map with where you have safe areas and where you have less safe and less safe and less safe and gradually act to make areas more and more safe, combining with people to expand the geographical area of safety. And Australia would be a great place to do that, right? It's an island, right? It's really, you know, you don't have to shut down a lot of the transportation in order to make it safe. So that sounds like a really smart idea.

To act aggressively. And it's only going to be for, you know, a month or two. You know, everyone else is going to die if they decide not to act. And you can also have flights back and forth to Wuhan, China in a week. And to Beijing. I mean, they're safe. So you can have all kinds of commerce with China and with South Korea in maybe three weeks.

So we rebuild the world out of the spaces that are safe so that's i think the right strategy now there's another thing that we are doing so a week ago today a week ago saturday night a week ago i sent out a note on twitter asking for people to volunteer to help stop the outbreak. And what are we doing? Well, basically, we're telling people what's happening, trying to make them aware of the severity of the disease, number one, and number two, what they can do about it. And what they can do about it includes, you know, sort of what they can do themselves, but obviously includes pounding on the doors of the, quote, decision makers who have control over the infrastructure or whatever it is and the police and so on in order to get their involvement in solving the problem. And it doesn't have to be at a national level. It can be at a state level or at a town level or whatever.

In China, when they shut down places, they shut down town borders. They had people at the border checking people and asking them, you know, did they have symptoms, where did they come from, and they couldn't come in unless they had the right credentials. So this is the creation of these safe spaces, but that requires some kind of organized action on the part of people. And organized action usually means government. The other places that can take action are our businesses. Right. So businesses can say everyone should work at home, just like, you know, the schools should stop.

Right. So people should work at home, just like, you know, the schools should stop, right? So people should work from home. So Apple today told employees to work it from home and Facebook and, and, um, um, what was it in Seattle? A couple of companies. I mean, Seattle is right now the worst place in the U S but just wait, New York is going to be worse in a couple of days. The point is that companies have a lot of power. They can make their employees feel comfortable staying at home and arrange, or they have a huge amount of economic power to help with making sure that people get tested. And all of those things matter right now in a huge way. So what we're trying to do is to have every possible organization, whether it's individuals, families, communities, businesses, governments, local, national, whatever it is, to influence them to take more aggressive action earlier, right? The sooner we take action, the less bad it'll be. If we knock down the transmission rate from 1.5 to 1.1, we'll have a little bit more time to stop it. But if we remember that we can stop it,

then we'll take the right approach. But what's happening right now is people are saying, hey, right, Elon Musk today says it's dumb to panic about coronavirus. And my colleague Nassim Wright said it's dumb to say that it's dumb to panic, right? Panic is a rational behavior when it causes people to act collectively to stop a problem. If an individual has, you know, two or three percent chance of dying or one percent chance of dying, they may say, hey, I'll take the risk. I don't know why, but maybe they'll take the risk. I don't think they should take the risk. I don't know why, but maybe they'll take the risk. I don't think they should take the risk personally. I also don't think they should take the risk of being in the hospital with an ICU with one in 10 chance or, or severely ill in hospital, 20% chance. And that includes the

fact that the hospitals are going to be overrun with people. So they probably won't get care, right? So their death rate in that case may go up tremendously, maybe to 20%. We don't know, right? It's a breakdown of society. But an individual may say, hey, it's one in 10 chance, I'll take my risk. But the society can't take that chance. And so the panic is a collective response where we all get together and say, let's act now, not later. The fact that there are 3,000 or whatever it is people dead today from this disease is not the point. The point is that it's growing exponentially. And in a month time, a month time, right, it's 10,000 times worse. And in two months time, it's 100 million times worse and In two months time it's a hundred million times worse And if we have a hundred cases today, which we have we have many more we have thousands of cases today

And that's the whole world. So we only have two months and if we act now we can stop it in three weeks And if we act next week, it'll take six week. Why should we wait? That when you sign go ahead the positive thing here is that The math can help us when we act and this is why It's so powerful to understand the math because remember I said that the most important thing is whether the R0 or the rate per day is greater than one or less than one. If it's greater than one, you're going to lose, right? Because it's growing. But if it's less than one,

it doesn't take much time for it to disappear. And it's gone. A few weeks, that's it. So the point is that the very fact of the exponential process, the multiplicative process, means that you're in real trouble if you're on the bad end of it. And you're fine if you're on the right end of it. So if we act now and R0, or the multiplication rate, is less than one, it's going to be gone. That's what happened in China. That's what's happening in Korea. So there's only a few people relatively speaking. Right.

It's still tragedy. Right. But there are a few people who get sick and die relative to the population of the world. But the way to get there. Is very simple. You act now, reduce are not less than one. We can discuss it later, right? If we overact now, what's the loss? We'll just lose the disease more quickly. So act now, overreact, and then, hey, you know, maybe a week from now we can ease up.

We don't have to wait for three weeks because we find that we don't need quite as aggressive a response. This is what I've been trying to communicate to people. It's not that it's not a question. This is such an important thing. There are people who are pessimistic and there are people who are optimistic. Both converge to doing nothing. Both converge, exactly. Both are about what the world is doing. What's going to happen? I'm going to sit here and watch. Is it going to be good or bad? The alternative is to make a choice

and to choose what world we're going to live in. And we have that ability to make that choice collectively. So there's a choice of life or there's a choice of death. We have to make that choice. And whether that choice is going to be made by governments or whether we're going to get together with our friends and make safe spaces, that's the second question. The first question is, are we making the choice to take action? That's the question.

I'm going to make sure we get this podcast conversation and many of your notes into the hands of some Australian politicians. But assuming our governments fail us, hopefully it doesn't come to that, you mentioned there are many things we can be doing as individuals. For people who want to join the fellowship of the doers, are there any guides you would direct them to any websites or sources is there anything they should be looking up on the web absolutely people who want to make it make a difference so there's a website that this happened in two days i can tell you we now have

i don't know well over 500 volunteers that are working on this, building websites, creating apps, doing communications, doing analysis and putting together important information and communicating it to all kinds of people. So we created a website called endcoronavirus.org. So remember the.org, it's endcoronavirus.org. And on that website, we have both information about what's happening and the guides. The guides are for individuals, communities and governments and for businesses and for families. And it explains that, you know, the most basic thing is kind of all aspects of social distancing, making sure that you're, you don't go to meetings if you don't have to, you know, do them virtually, that you, if you have to go out in public, you are careful about touching surfaces, you know, door handles and, you know, all kinds of things that, you know, elevator buttons. And so, you know, either use gloves or use paper towels or you wash and you surely wash your hands frequently because surfaces get contaminated by the virus and you pick it up from other people that have touched those surfaces.

So there's a basic rules of social distancing and not going to meetings, canceling large meetings. Find other ways to do business. And, you know, there are businesses that can send people home and, shut down the schools and so kids will get a vacation for a week I mean tragedy right I mean that's not the problem here let them read the encyclopedia I did that for a couple of weeks when I was in elementary school when I didn't like what was going on in the school so so find good ways for people to engage and improve their lives while we're safe from transmission. So there's the basic level of the individual. But the other thing is if you want to help us, join our volunteers.

I mean, this is the action that we need. I mean, you can't imagine how capable and incredibly active the people who are doing this. I mean, there's so many things going on. And they're figuring out what to do. This is the self-organization that we were talking about. This is the ability to create collective behaviors. And the main thing is that it's not about telling people what to do. It's about them understanding what we're trying to do and to make it happen. And what we really want to do, and that's the mission, is to stop the outbreak. And we can, right? That's the point. We can if we do it together.

So join, and let's make it happen. If you go to the website right now, there is an email address that people can send a note to. It's just NECSI volunteers. So NECSI are the initials of the New England Complex Systems Institute. So NECSI volunteers at gmail.com. And I'll put links to all of this in my show notes as well. And again, the website is ncoronavirus.org. And we will have a sign up also for volunteers and for people to just participate in the activity

of stopping this outbreak. You nearby, Yam, this has been the most enthralling and important podcast episode I've recorded to date. Thank you so much for your time and for sharing your insights. Thank you. I really appreciate the opportunity and I really appreciate that people will hear about this and join us. That's so important right now. Be well. Be well, everyone.

Thank you so much for listening. I hope you took as much from that conversation as I did. During the recording, it was announced that Italy imposed sweeping lockdown measures on Lombardy. Yanir and I celebrated that with some fist pumps. But of course, there is so much more to do. We have to protect the species, or at least our local communities. To links to everything Yanir mentioned, including how you can join the fight to end coronavirus, you'll find those in the show notes to this podcast, which you can also find on my website. That's www.josephnoelwalker.com. To spell that out, www.josephnoelwalker.com. Until next time, thank you for listening and take care of yourselves. Ciao.