What does it mean to think? What are the traits of thinking systems that we could use to identify them? Different environmental variables call for different strategies in individual and collective cognition — what defines the threshold at which so-called “solid” brains transition into “liquids”? And how might we apply these and related lessons from ecology and evolution to help steward a diverse and thriving future with technology, and keep the biosphere afloat?
Welcome to COMPLEXITY, the official podcast of the Santa Fe Institute. I’m your host, Michael Garfield, and every other week we’ll bring you with us for far-ranging conversations with our worldwide network of rigorous researchers developing new frameworks to explain the deepest mysteries of the universe.
This week on the show we talk to SFI External Professor Ricard Solé of the Universitat Pompeu Fabra (Website, Twitter, Google Scholar) about liquid and solid brains, the scaling of cognition, criticality, contagions, and terraforming our own planet with synthetic bio.
Be sure to check out our extensive show notes with links to all our references at complexity.simplecast.com. If you value our research and communication efforts, please subscribe, rate and review us at Apple Podcasts or Spotify, and consider making a donation — or finding other ways to engage with us, including our upcoming program for Undergraduate Complexity Research, our new SFI Press book Ex Machina by John H. Miller, and an open postdoctoral fellowship in Belief Dynamics — at santafe.edu/engage.
Lastly, join us June 19-23 for Collective Intelligence: Foundations + Radical Ideas, a first-ever event open to both academics and professionals, with sessions on adaptive matter, animal groups, brains, AI, teams, and more. Space is limited! Apps close February 1st. Learn more on our website.
Thank you for listening!
Join our Facebook discussion group to meet like minds and talk about each episode.
Podcast theme music by Mitch Mignano.
Follow us on social media:
Twitter • YouTube • Facebook • Instagram • LinkedIn
Referenced & Related Works
Liquid and Solid Brains: Mapping the Cognition Space
SFI Seminar by Ricard Solé
John Hopfield (re: biology as computation)
Synthetic transitions: towards a new synthesis
by Ricard Solé
Complexity 93 - Kate Adamala on Synthetic Biology, Origins of Life, and Bioethics
The Multiple Paths to Multiple Life
by Chris Kempes and David Krakauer
Simon Conway Morris (re: macroevolutionary trends)
Scale and information-processing thresholds in Holocene social evolution
by Jaewon Shin et al.
Smarter Parts Make Collective Systems Too Stubborn
by Jordana Cepelewicz at Quanta Magazine
Complexity 90 - Caleb Scharf on The Ascent of Information: Life in The Human Dataome
Will Ratcliff (re: yeasts and emergent multi-cellularity)
Synthetic criticality in cellular brains
by Ricard Solé et al.
Tom Ray (re: artificial life)
Complexity and fragility in ecological networks
by Ricard Solé and José Montoya
Ecological Networks and Their Fragility
by José Montoya, Stuart Pimm, and Ricard Solé
The small world of human language
by Ramon Ferrer i Cancho and Ricard Solé
Macroscopic patterns of interacting contagions are indistinguishable from social reinforcement
by Laurent Hébert-Dufresne, Sam Scarpino, and Jean-Gabriel Young
Complexity 56 - J. Doyne Farmer on The Complexity Economics Revolution
Complexity 66 - Katherine Collins on Better Investing Through Biomimicry
Chris Langton (re: criticality)
Jim Crutchfield (re: the edge of chaos)
Per Bak (re: self-organized criticality)
Complexity 10 - Melanie Moses on Metabolic Scaling in Biology & Computation
Complexity 3 - Sabine Hauert on Swarming Across Scales
Niles Eldredge (re: punctuated equilibria)
Terraforming the biosphere: can bioengineering save us?
SFI Seminar by Ricard Solé
Ecological complexity and the biosphere: the next 30 years
by Ricard Solé and Simon Levin
Ecological firewalls for synthetic biology
by Blai Vidiella and Ricard Solé
Rachel Armstrong (re: synthetic biology for CO2 fixing in concrete)
Stewardship of global collective behavior
by Joseph Bak-Coleman et al.
Complexity 64 - Reconstructing Ancient Superhighways with Stefani Crabtree and Devin White
Complexity 5 - Jennifer Dunne on Food Webs & ArchaeoEcology
Ricard Solé:
If you reduce the uncertainty of the world, if in a way you predict what's going to happen with the environment, the payoff of this is very large. Now, how do you manage to do that? In ant colonies you have a system that needs to solve problems that have to do with maintaining the colony and eventually reproducing the colony somewhere else. For that, you need some intelligence that is able to gather information from the environment, probably to send to that environment in a special way. So you need to know where the resources are. That probably immediately creates the need for a distributed set of agents that can facilitate that process. So you have the system – the little brains that form a big brain.
Then it's us. The solid brains where, we know from theory at least, that because neurons are located in specific positions, but have connections that learn that well, that participate in the learning process. You have almost an infinite landscape of possible states. So that gives us in a single individual the possibility of actually having a pretty complex decision-making system.
SFI/Michael Garfield:
What does it mean to think, what are the traits of thinking systems that we could use to identify them? Different environmental variables call for different strategies in individual and collective cognition. What defines the threshold at which so-called solid brains transition into liquids? And how might we apply these in related lessons from ecology and evolution to help steward a diverse and thriving future with technology and keep the biosphere afloat? Welcome to Complexity, the official podcast of the Santa Fe Institute. I'm your host, Michael Garfield, and every other week we'll bring you with us for far-ranging conversations with our worldwide network of rigorous researchers developing new frameworks to explain the deepest mysteries of the universe.
This week on the show we talked to SFI External Professor Ricard Solé of the Universitat Pompeu Fabra about liquid and solid brains, the scaling of cognition, criticality, contagions and terraforming our own planet with synthetic bio. Be sure to check out our extensive show notes with links to all our references at complexity.simplecast.com. If you value our research and communication efforts, please subscribe, rate, and review us at Apple Podcasts or Spotify and consider making a donation or finding other ways to engage with us, including our upcoming program for undergraduate complexity research, our new SFI press book, Ex Machina by SFI External Professor John Miller, and an open post-doctoral fellowship in belief dynamics at santafe.edu/engage.
Lastly, join us June 19th through the 23rd for collective intelligence foundations and radical ideas. First ever event, open to both academics and professionals with sessions on adaptive matter, animal groups, brains, AI teams, and more space is limited. Applications close on February 1. Learn more on our website. We'll be taking a short break over the holidays. We'll be back the second week of January. Thanks so much for listening and happy holidays.
All right, Ricard Solé, welcome to Complexity podcast.
Ricard Solé:
Thank you.
SFI/Michael Garfield:
So let's start with a bit of backstory to anchor the work that you're doing in a history of your own intellectual curiosity. What motivated you to become a scientist in the first place and what drew you into the kinds of research questions that you pursue in your work?
Ricard Solé:
When I was a kid, I liked nature a lot. You could say I was kind of a naturalist. I discovered there was something called the theory of evolution when I was around 10, and it had kind of a big impact -- like saying, wow, there's a lot of logic here in life. Then when I was in high school, I accidently had contact with complexity in a book that had been translated to Spanish from a very classic work on towers of theoretical biology with chapters by [Conrad H.] Waddington, and by [Stuart] Kaufman. All of a sudden I realized that there was a connection between mathematics, which I kind of liked, and things like brain development – what happens inside cells -- and evolution.
I became very curious and I kept wanting to be a biologist. I hated physics. But in the last year before university, I had this amazing teacher who gave us a course in physics and I fell in love. So I decided to do both biology and physics, which was a very good combination and brought me into questions of complex systems, of course.
SFI/Michael Garfield:
So that's actually an interesting dichotomy that you explored in the talk that you just gave here at SFI on liquid and solid brains about the difference between biology and physics and how that difference is anchored in cognition. You know, that's going to take us on a tangent, but I'd actually love to hear you talk a little bit about that just as a way of kind of framing this stuff, you know, in terms of how we're thinking about things that are thinking.
Ricard Solé:
Yes, there are a lot of interesting problems connected to that. I mentioned this paper by John Hopfield where he makes explicit mention that if we want to find out what makes biology different from physics, probably computation is one of the core ideas -- that biological entities might have a lot of things that you can describe in terms of physics, but they have this special character: they take information from the environment, they process it, and respond in adaptive ways and that's a very distinct phenomenon. But the interesting thing is that it brings in novel questions like, for example, at the point all of this starts, where an agent is in a prebiotic or early evolution scenario, when is information relevant and how do you identify the nature of information? And is there a distinction between information and computation. These are all open questions.
SFI/Michael Garfield:
Maybe the right place to start then would be this paper that you wrote, “Synthetic Transitions towards a New Synthesis,” where you're talking about something we've discussed a couple times on the show, including with Kate Adamala, and it certainly comes up in “The Multiple Paths to Multiple Life” paper that Chris Kempes and David Krakauer wrote together and held a workshop around recently. These questions of how do we get at what it means to be alive, to think of it in kind of like a functional sense, rather than in terms of specific material instances or substrates. But then again, in your paper, one of the things that you talk about is the way that living systems early on in their evolution seemed to have depended on environmental scaffolding and the way that some of these processes were outside of the system before they were internalized, before they were made part of the default toolkit of what we think of as a living thing.
So I'd love to hear you unpack this paper a little bit, because the question of scaffolding came up again in your talk recently, where you were describing how it seems that in ant colonies, the vast majority of them form nests even in something that we might think of as like a liquid or a fluid brain and that there is this sort of pushing against or utilizing of a kind of a static component.
Ricard Solé:
The paper was an ambitious one because what I wanted to bring to it is the idea that we can look back into the major revolution transitions from the synthetic perspective, whether or not we can actually build the transitions -- in relation with, for example, protocells. Clearly very important problems that are related as you were mentioning -- you have a scaffolding, you have embodiment, things that very often in the literature are pretty much ignored. People that kind of make models of protocells that somehow replicate –but somehow it's very important. I mean, you have to get there against the physics of equilibrium that favors closed membranes, but it's not keen to make them destabilize, which is connected with the fact you have to replicate. And that again connects with other problems, like, for example, when is this something that gets into the control of some kind of information?
Because the previous scenarios that probably predate the origin of life -- the first cells -- are not made of cells that replicate in a very reliable way. And I even doubt that we should look so closely all the time into vesicles because soft matter has so many configurations and maybe anything that predates these embodied protocells may have been totally different, maybe forms with different molecules kind of helping each other to replicate, but still not as an agent. There's no agent really there. And we need to put physics there. We need to put the physics of how you make things destabilize and how is this connected with Darwinian evolution, which seem to be two very different distinct things. But we need to make a connection. And a particular embodiment is important --embodiment because agents in living systems typically interact with an external world and sometimes they engineer.
Ant colonists engineer their material world. Somebody said that they take advantage of dirt, which is a very malleable thing and that's part of their success. We humans, on the other hand, have also engineered the biosphere and it's interesting that in both cases we have an intelligence, but again, the nature of the intelligence is different. One is liquid and another's solid, and both conquered the planet.
SFI/Michael Garfield:
So to that point, you mentioned Simon Conway Morris's work in your talk recently and he's this longstanding, very vocal proponent of the role of constraints and the importance of convergence and evolutionary processes. And for anybody who's been listening to the show, that's a pretty big thing here at SFI. You look at the work of Geoffrey West and Chris Kempes and you see this notion that there's this enormous region of potential diversity that never gets touched because of the importance of those constraints. You spoke about this in terms of theorizing a kind of morpho-space of possible cognitive structures.
Part of this question of when major evolutionary transitions happen and why some of them are more common than others is something I'd love to hear you link to some comments that you make here. Something else that I've heard Conway Morris at Cambridge talk about, which is how often earlier instances of life forms seem remarkably more complicated and inefficient if you look at jellyfish genome and how it's enormous and so people expected the human genome to be even bigger and it turned out it was far smaller, far more parsimonious. And then that seems related to a comment you make later about early theoretical arguments indicating that there's a maximum length for RNA chains scaling as the inverse of mutation rates that are beyond this theoretical maximums, whereby a system experiences a so-called error catastrophe, a phase transition where genetic information is lost.
Again, maybe I'm trying to fold it all too much into one thing, but when I hear you comparing all of these different kinds of brains, one of the things that comes up is the work that was done here on the diastol and systolic activity of social scaling and informational scaling like Hajime Shamma and David Wolpert and Tim Kohler and a few others wrote that paper where they were talking about how society in the holoscence gets to a point where it can't manage its own informational organization and has to come up with a new structure. And so like you look through history and you see these kinds of libraries, their cultural technologies, emerge as a way of managing the coordination of all of this stuff.
I probably just bit off way more than I could chew, but I'd love to hear you kind of dive into that and take it wherever you see is appropriate.
Ricard Solé:
Well, on the one hand, it’s very clear that it took a while to understand that why very large genomes might not necessarily go into very extraordinary complexity. The jellyfish example, I think, is a very good one, but you find this also in plants. We do know that with complex systems, people have clearly good intuition about it all – that the network is what's really important. The interactions are the things that create diversity of states and ways of connecting those states. I think that's what one clear thing. On the other hand, you mentioned the space of cognition, which is one of my interests, and whether or not we can make a map of the cognitive space.
And it's also interesting when thinking about humans. I had this conversation recently with Michael Lachmann and he was making a very good point about how we'll say humans are very intelligent, but of course the cultural dimension is so important. He argued that if you isolate a human from language, from culture, then they’re nothing. They have this huge brain that is essentially disconnected in a way and you will say this is a stupid agent because it doesn't do anything really interesting. So in a way that's why I think we are kind of a corner in this [cognition?] space, which on the other hand has big empty spaces and we need to figure out why.
SFI/Michael Garfield:
That seems related to something that you mentioned in this talk. I always appreciate it when people bring in cultural references, to anchor this stuff and, and you brought up the Borg, why is it that we don't see the Borg in nature? You suggested it has to do with these trade-offs between individual and collective cognition. There's that research that was talking about how the longer the memory of agents in a system is, the more stubborn that system is generally to updating. And so the article was called “Smarter Parts Make Collective Systems Too Stubborn.” But that seems again to kind of get at one way of thinking about the intelligence as something that's happening across scales.
I talked with Caleb Scharf about the way that we outboard cognition into what he calls the “dataome” into all of our external computational, the extended phenotype of humanity. That we we're becoming in certain ways more specialized or more dependent on one another. Like you said, the individual person is pretty helpless. And so I'm curious if this line of thinking gets to the question that you ask in this synthetic transitions paper about why it is that some of these evolutionary transitions are so rare and the kind of related question that you asked in your talk recently about why it is that, like you said just a moment ago, humans appear to be such outliers in this face space of cognitive possibility, that we're extending ourselves into the environment so much more than other creatures like us.
Ricard Solé:
Yes, I think that on the one hand is the idea of [plotting?] this space of cognition, because we really need to build a theory of why different cognitions, why some cognitions that we might imagine are not there. One of the things that I think it's important to have in mind is that when you look at this space, you have these big chunks of empty solutions, which doesn't mean that they're not available. Maybe we can engineer them, but the evolution may have a hard time. But when you look in particular to humans, it's an interesting situation because, as I said before, it looks like kind of escapes from the general rule. For general rule I mean something that we still need to figure out, which is this: the complexity drain, as I mentioned in the talk.
For example, when I was in Panama some years ago, I could see these army ants that move through the forest like a single organism. They are really amazing the way they are plastic and managing their environment. But the individuals are essentially very dumb. I mean they are blind, they communicate with chemical signals. And in December in the forest there's another colony of huge ants like with big eyes, small colony. And it was interesting to look at them because it was like each one was more or less on their own as if, okay, I have enough equipment to kind of make my own decisions. So this seems to be a trade-off -- the trade-off that we see multicellularity, many of our cells [ourselves?] really lost a lot of potential because they have to be specialized. But again we need to develop a theory and put ourselves within this evolutionary picture.
SFI/Michael Garfield:
So that's a good place I think for us to kind of fold this all together because you know, you talk about this specifically this transition into multicellularity in this piece and you know we've already talked about the environmental scaffolding and the specialization piece and I'm thinking about Will Ratcliff at Georgia Tech and how he's shown with the emergence of multicellularity in yeasts being a kind of a product of these physical relationships between the cells and where they are growing. And then we've got what you talk about, which is a minimal form of multicellularity comprising of persister cells associated with cell subpopulations that spontaneously switch back and forth against multiple resistant phenotypes as a bet hedging strategy. This reminded me of what David Krakauer and I talked about back in episode 29 about when in systems you see conditions favoring specialists versus generalists. So I'd love to hear you reach back into multicellularity and synthesize all of that.
Ricard Solé:
Some people said that this is a transition that's one of the easy ones because in a way we have been finding out over the years that being together could be helpful. Maybe because you have a resource there and it's good to be together because we want to. If you attach to each other, we'll keep around a food source and exploit it. But of course you can bring in the idea that it's more costly and that's why Will Ratcliff’s work on synthetic multicellularity has been a groundbreaking finding. Basically it shows that under conditions that are easy to make in the lab, you can actually see the emergence of an entity that is a kind of a proto-multicellularity in the sense that you don't have what we have in the modern multicellularity, which is self-specialized into becoming spores or something that allows for building a whole thing. Here is more like you grow and you can break.
But again, we need to understand what the conditions that predate finally multicellularity are. You were mentioning this work that we did from some years ago -- it's a modeling work -- but it was pretty much an interesting result. That if you put very simple things at work, cells that can attach to each other, and one just takes something from the environment and the other specializes in removing a toxic from the environment, a stressors. From simple conditions we ended up into something we call a proto-organism, which was that these cells evolve addition and they put together in a way that when you observe the final result, this is like an organism because you have an internal and an external state. So it's an external wall organized in very ordered ways to deal with that. So again, you don't have multicellularity as we know because there's no genetic control. The process is not yet the life cycles, which is this really big question you have to solve. But you do have something that reminds you a lot of an organism and maybe that was part of the landscape that predated the major transition
SFI/Michael Garfield:
Now seems like a good time to just pay our dues to precedents for this work in A-Life. You bring up Tom Ray and Tierra in this and in particular in the context of the acceptance of the possibility that sort of parasitic forms might be an ineradicable dimension of these kinds of systems. I'd love to hear you talk a little bit about how that links into all of this.
Ricard Solé:
Yes. One of the things that I do believe, but we don't yet have a complete theory for, is that it seems that parasites are an inevitable part of reality and in all scales, in fact. In Ray's work, one of the nice results among others was that when he put this digital biosphere in the computer, which was kind of a very early attempt of doing that, he found out that one of the first things that happens is that you evolve your agents that aren't able to replicate by themselves but they can replicate by using pieces of code from the other ones, suggesting, and I think that is a very generic result, that parasites are inevitable.
I think that this has two consequences. One is that it explains that parasites are everywhere. It's probably something that is part of the opportunities of nature for cheating in particular. But also I think that when we discuss life somewhere else or alternative life, this could happen Sometimes we say that maybe what is generated by evolution is very path dependent, but they also believe that they are fundamental parts of the logic like parasites that are inevitable here or and elsewhere.
SFI/Michael Garfield:
One of the things that you talk about in your paper is how, since the advent of computer viruses, they have “evolved towards more silent, apparently harmless designs based on their potential to ‘integrate’ themselves within the host machines, where they remain undetected,” which is very sly. This seems like a good place to peg into the paper that you co-authored with José Montoya and Stuart Pimm in Nature on “Ecological Networks and their Fragility.” If you can just introduce us to this way of thinking about the difference between the networks that we observe kind of generally and the dynamics that exist in ecological systems and why those things are different.
One of the things that you talk about here is how economies appear to have a kind of rich get richer dynamic and that when we look into, for example, a rainforest, those networks don't display the same kind of benefits. If you could, just give us a bit of background on thinking in this way and then what your wok reveals to us about the way that ecosystems are distinct and why it is that we see things coming into balance -- like viruses evolving to not destroy their host organisms.
Ricard Solé:
Although we live still in the wake of the first years of this century when networks became a revolution, which has been good for us complex systems scientists, network thinking has been an ecology since the fifties. And it has been a system science almost forever since it was constituted as a field, and network thinking has been around for a long time and shaping a lot of our understanding of natural systems because one thing that came about very early in the seventies with the war by Robert May and others is that even though you can find ecosystems that are very different, there seems to be clear general laws -- like the relationship between the number of species and connectivity. You can increase connectivity, but if you do that then you make the system more unstable, so you reduce diversity. So it's a clear dynamical trade off.
We find the same thing if we go into the rainforest and, for example, sample all the trees in a chunk of the forest. You'll see that most species are rare and a few of them are really dominant. But then you move into another place and you make the same kind of work and you find out that you have exactly the same distribution but the species composition is different. Which is saying that, on the one hand, there are very strong laws of organization that have to do with very dynamical states, but at the same time there is this path dependence -- that the exact way in which you organize the system depends on the path story. This has been a guiding principle.
The network principles that came out with small-world and scale-free networks, etc., provided something that we hadn't explored before, which is the fragility of the system to extinctions. Usually we think of extinctions in terms of, for instance, a given species gets extinct for some reason. But the system is interconnected. You can have cascades of extinction and these reveal early that future scenarios of extinction have become this because when you lose some species you can trigger events that accelerate the whole process.
SFI/Michael Garfield:
You talk in this paper about compartments in ecological networks and how they correspond and sometimes don't to the boundaries between habitats or the feeding relationships between organisms. The word “compartments” comes up again in your paper on synthetic transitions. You spoke earlier about the formations of vesicles in informational boundaries around organisms. I'd like to hear a bit about how you see there being a kind of a general insight there in how the way that we can think about compartments in an actual organism and the compartments in an ecosystem (or it may not be different) and whether similar forces are driving that -- why we don't see compartments sometimes.
Ricard Solé:
Well it's a difficult question in one sense that the presence of, in a way, modular structures in biology can be because you have an embodiment. So you have something that defines a boundary and that automatically defines the inside where you can then organize other models. Then if you want to respond to questions such as when the real cells emerge and where, and the reliable systems able to explain the real selection. When you go up into the ecosystem level, you do observe all kinds of organization. Its ecological networks have this nested structure. On the other hand, there are species that play an especially relevant role – with different layers if you want a hierarchical organization. But in this case these models or nested structures are not made of embodied objects. It's more like the dynamical rules that provide the right stability to favor, for example, particular models in the system, which happens also inside cells for gene regulatory networks, for example.
SFI/Michael Garfield:
I’d like to try to link this to a piece you wrote with Ramon Ferrer i Cancho on the small world of human language. You mentioned this term in passing just a moment ago. For those who are not familiar with small world networks, I'd love to hear you explain how this work can kind of be generalized with the work on ecological networks. And then from there, I have a question for you about modularity and the relationship between linguistic and biological diversity. There’s this everyone-can-plug-into-a-tower-of-babel archetype or myth, and we can look at things like the way that it seems that the worldwide web is not unifying in exactly the same way as it was anticipated to in the nineties and the way that we're seeing this fractious or entropic effect where these large systems are kind of breaking down into modular components. I look at stuff like Sam Scarpino, Laurent Hébert-Dufresne, and Ben Althouse’s work and folks who have been working on contagions and not just biological contagions but contagions of belief and behavior. I guess this question is ultimately about the nature of biodiversity and of food web relationships as an adaptive strategy to encourage robustness and resilience and to prevent catastrophic outcomes.
Ricard Solé:
Well, you give me a hard time really, with these questions. In relation with the small world of female language, that was one of the most exciting things we did in the lab. Ramon came with idea that we're not thinking in the simplest way of building a network, which is to take two words as connected if they appear one after the other, inside one sentence at least. And from that very simple idea, we built these huge networks from British [National] Corpus, for example. And we found out two amazing things. One was that the system was small world, you navigate into a repertoire of really large number of words.
I mean even people might not think about that maybe sharing 50,000 words and maybe more except that we don't use most of them, most of the time. That excludes of course professional football players. But the thing is that we find out that despite having a corpus of you know, thousands and thousands of words, the distance you have to jump in terms of going from one word to another in the network was less than three. So it was really easy to go from one part to the other. Then there were some words that were extremely connected, words like “the,” preposition [article??] for example, that you will say, well they don't have much content, but what we found and was confirmed by other people at the same time working on semantic networks is this beautiful result that yes, okay, you have this architecture etc. and was the big lesson here.
Well, you can navigate very quickly through the network and that in a way it doesn't explain, but it provides some understanding of why we are building sentences when we talk so quickly, having in mind that if you think in the synthetic structure you maybe should stop thinking each time who comes after the other. But you have these [apps? Hubs?]] that make navigation so quickly. It's probably some of the big advantages of the complex language that we use, which is ambiguous, but ambiguity again is connected with the hubs -- the hubs that allow you to run away, run quickly. Of course this opens other questions, which is what is the network and how is this connected with the neuroscience part of the story, which we don't know.
And you mentioned diversity and language and ecosystems and it's been really clear from a very long time ago that we share a number of common things in terms of how languages emerge, how diversity of languages is sustained over time and even how they are connected. I mean the places on the planet where you have most diversity are the places where you have more diversity of languages. Again, they face the same problems: they are becoming extinct and extinct quite quickly, but for reasons that are different. Languages that are very large and dominant they tend to kind of observe the speakers of the other languages and so that's the main path to extinction, whereas of course species suffer our direct pressure and to nature.
SFI/Michael Garfield:
So actually that reminds me to double back to what it is about the nature of ecological networks that inhibits the sort of rich get richer phenomenon that we see in economics, where nodes with more connections kind of draw or attract more and more connections to themselves. Because it seems like when I'm working on this show and I'm having conversations with SFI economists and I'm thinking about the emergent political economies program, then on the one hand we've got people like Doyne Farmer that are thinking in terms of market ecologies and are applying ecological thinking to this stuff. On the other hand, it seems relatively clear that at least at the timescale that matters to us as human beings that we're thinking about and behaving in the economy in ways that are ill-informed by these kinds of ecological dynamics.
I look to Katherine Collins' episode and her plea for us to think about this stuff in more of a biomimetic way, to draw more lessons from ecological systems into economic activity. Is it the case, then, in economics that the rich really do get richer or are we just looking at it on such a narrow time scale that we're missing the fact that ultimately these systems are going to sort of balance themselves out again in a way that makes human beings as super generalists a little bit more modest in the way that we relate to the rest of the biosphere?
Ricard Solé:
Clearly they are very general rules in terms of the dynamics that apply for both two ecology and economics,. You identify competition, you can identify corporation [cooperation?], you see that diversity plays a role in the economy to make the economy more dynamic. But of course this is a very different context in terms of what the payoffs for things are. In the real economy, clearly the rich get richer in the worst sense and that particular individuals or particular companies get all the profits. And then in fact when we were working on [???] many years ago, I remember that one of the obvious things that we observed when looking at examples of this phenomenon was that the economy is probably moving all the time into this kind of unstable state, and one of the predictions clearly from the model is the Zipf law of wages and everything. Because if you don't control the system, you get into that state. In ecology, with this kind of increasing returns in ecology, I think the picture is different because you do also have these power distributions for biomass, for example. And you have also power distributions for species abundance, as was mentioned with the rainforest, and in marine habitat everywhere. So again, you have these universals. But even though there are some species that are keystone that are maybe, for example, ecological engineers, they have a really important role in controlling energy matter. The turnover is very large.
So as I was saying, there's a very dynamical process that also in the long run involves Red Queen effects.
SFI/Michael Garfield:
Talk a little bit more about that concept, because not everyone may be familiar with it.
Ricard Solé:
The Red Queen is this idea that was formulated many years ago by Leigh Van Valen, where he said that when you look at natural systems, you tend to think of species as something separated from the environment. You evolve because there's an environment you have to adapt to. But in fact every single species has to change, also, in relation to the rest of the ecological network. The simple example is if I'm a prey, and if my predator gains some advantage in detecting me, I have to evolve. I have to co-evolve and I do something else, right? Maybe be less visible or be smaller. And this [arms?] race happens in all the networks. One of the consequences of that is that you should expect every species to become extinct sooner or later.
So every single species, everyone, in the long run, becomes extinct. And those kinds of dynamics are kind of an illustration that even if you may think that one species is now kind of a winner, this is just a transient phenomenon. Also, when you go into the real world, maybe the winner in one place is a different one from the winner in another place. So in that respect heterogeneity and the role of space in ecological dynamics makes a big difference because economy also has become a very global phenomenon and all of this has kind of been erased.
SFI/Michael Garfield:
It strikes me that we've been dancing around a key concept this entire time that I think we ought to just make explicit here, which is the concept of criticality and the actual state of the connectedness of these systems -- and how that relates to the phase transition that you explore in your work between liquid brains and solid brains and how it relates to these major evolutionary transitions. So if you can break that down for us, I'd appreciate it.
Ricard Solé:
Yes, criticality is a concept that has been becoming more and more central in many ways. In the nineties it became a source of excitement because there was the idea, from Chris Langton, that systems that perform computations might benefit from being close to a critical state, a state where you have order, on the one hand, which you need for memory, but this is already what you need to rearrange information and manipulate it. And then at the same time, Jim Crutchfield came with idea that there was this age of chaos again for systems and if you measure the computation in a system that is at the edge of order and disorder, you find out that almost every measure that you think can define computation has a pick [peak?].
And also at the same time Per Bak came with the idea of civil organized criticality -- the idea that you can actually evolve into that state. This idea has been shaping a lot over the years and it's been interesting in the last years where we have a lot of data, a lot of very good information that criticality or very close to criticality states seem to be generic in the brain. So the brain actually uses, at least in terms of dynamical state, it sleeps close to a critical state [??] viruses live on the error catastrophe, right? Which is the transition from the order and disorder or the meaning that you keep information, you experience selection and disorder, and that you're unable to store information anymore. Essentially you will go into extinction.
The same seems to happen in unstable cancer and in behavior in the way flocks of birds are able to manage to be so flexible. You actually see kind of good intuition, you see order and disorder, and when you measure that precisely -- the way they interact -- and you use tools from statistical physics, you realize they live in the critical state. So being in the critical state, at least for cognition, is something that prepares you to respond very quickly while you have order inside, right? It's not a response that is totally disordered but one that provides a lot of flexibility. One thing that is open is that you think in the brain of course that you have this very dynamical state, but when you think in terms of the condition that is performed by the brain, how is this connected?
Is this just something that is kind of an overall dynamical state that is useful for going fast into responding or is it something deeper which might even constrain the way we organize our minds?
SFI/Michael Garfield:
This gets to a question that I was able to explore a little bit with Melanie Moses way back in episode 10 -- talking about when evolutionary conditions favor a liquid brain like a colonial organism, like an ant colony, or when you can actually scale individual organisms up to something the size of a human being or an elephant without it breaking apart. And how that is related to when we had Deborah Gordon on the show and she was talking about the cognition of ant colonies. This has been kind of an ongoing theme about responsivity and the inherent pace of the environment -- the pace at which a system has to adapt to it.
In episode 29, David Krakauer talked about the coronavirus as having an identity that was more like a cloud in space rather than a point in the way that we typically think of a species. And I've heard a number of scholars from the humanities using this as an analogy for the conditions of modern human psychology, navigating the complexity and pace of modern life. There's this whole question for me when thinking about these conditions when a swarm versus when a self. To get back to a synthetic bio question, do you think that this helps inform us about things like where are the use cases that we would actually design for one strategy over another in order to achieve specific results?
Ricard Solé:
Yeah, that's something I want to know. One thing that I will say is that if you look for a general principle that is kind of an umbrella for the solid and the liquid as complex systems, I think it’s clearly a kind of a driving principle here -- why develop complexity, especially cognitive complexity. Well, because having a brain is a costly thing, having intelligence in general maybe. Because you need to maintain a system that is able to process information in a way that involves more than just sensing and responding. But
if you reduce the uncertainty of the world, if in a way you predict what's going to happen with the environment, the payoff is very large.
Now, how do you manage to do that? In ant colonies you have a system that needs to solve problems that have to do with maintaining the colony and eventually reproducing the colony somewhere else. For that, you need some intelligence that is able to gather information from the environment, probably to send to that environment in a special way. So you need to know where the resources are. That probably immediately creates the need for a distributed set of agents that can facilitate that process. So you have the system – the little brains that form a big brain.
Then it's us -- the solid brains where, we know from theory at least, that because neurons are located in specific positions, but have connections that learn that well, that participate in the learning process. You have almost an infinite landscape of possible states. So that gives us in a single individual the possibility of actually having a pretty complex decision-making system.
So you see kind of two different ways of going into that principle, managing information, reducing the uncertainty. The question is, is there any other intermediate state that we might create in the lab? Because it seems so far that there are the two solutions. But again, we need to be careful because when we look at biology, what we always find out are exceptions and that there are layers. Like [ver...??], which is this huge cell. Some people don't like to call it a cell, but this is a single entity with thousands of nuclei inside which behave as a single organism that changes shape as it searches in the environment. It's kind of an alien organism and eventually the big thing is for it to find out which place in the space around it has the most interesting nutrients. Then it changes its shape to actually take advantage of that. So the computation is the shape, which is kind of a really different picture from the standard cognition, right? So that's what in a way makes me think that maybe some evolutionary paths to things that we haven't seen were not possible, but that doesn't mean we cannot maybe build those solutions.
SFI/Michael Garfield:
So that seems to link directly to comments I've heard David Krakauer make about the possibility of designing dynamic national constitutions or organizational charters that respond to variations in the pace of their environment and the demands on cognitive load by basically folding and unfolding and having shorter or longer regulatory sequences depending on the needs of the moment.
But the comments you just made make me wonder if there is not some kind of inherent obstacle to that in evolution that has made it rare and kind of difficult to identify or to comprehend these possibilities. To drop another pop culture reference, I think most people who grew up or lived through the eighties remember Voltron and these notions of people writing these robots coming together into these enormous multicellular type robots. And there’s people like Sabine Hauert working in robotics labs who are doing these things where you're kind of playing around with this notion of a biphasic structure – one that can go back and forth between states, coming together as a single individual when it wants to achieve a particular task. You see this also in agile thinking and organizational dynamics, this notion of the teams that sort of self-assemble in order to achieve a particular project and then everybody retreats back into the woodwork to work on their own stuff. So at least it seems like a lot of people are thinking about this in kind of convergent ways and finding ways to deploy it.
But there are these challenges akin to the challenges faced by generalists in a really mature ecology, which are like challenges of the efficiency of systems like this, and I’d love to hear you speak to that.
Ricard Solé:
Yes, I think that at least in the context of robots – I love that you bring this up – I always thought that the understanding of technology (from the work by Doyne Farmer and Niles Eldredge a while ago) is kind of an experiment of evolution, a [parallel??] experiment where you brought in the virus, the computer viruses. And it's interesting that computer viruses of course have been created by humans and so there's no real natural selection, not things like mutation, but we have managed to introduce all of this as we face challenges with the computer virus, which from our observation follows kind of the path of natural viruses -- from just infecting to be hiding.
So technology offers a lot of opportunities to explore these things and you mentioned robots and I think it's interesting that the liquid solid part of the story could be seen in the robotic arenas -- swarm robotics and humanoid robots. There's a lot of work on these two areas and I can't avoid thinking how it's interesting that humans could be doing anything else, right? Anything that is not an agent that is very complex or a system that is made of small robots, simple robots that collectively do interesting things as if the engineering approach to that brings again the solid and the liquid -- why not totally different kind of agent architecture organization. And that brings the convergent idea that maybe technology also is discovering that is only essentially the solutions.
SFI/Michael Garfield:
After your talk the other day, Carlos Gershenson asked you about gaseous brains and that's actually something that one of our followers on Twitter asked me to ask you about because it's obviously an analogy that suggests a third state. And I would love for you to dig back into the answer that you gave Carlos about what are we even looking for and is it even possible to have this third state cognitive architecture?
Ricard Solé:
The short answer is a conjecture, of course. We define liquid as something that reshapes itself within a given container. Isn’t it in a way what happens with ants, for example, that they have the scaffolding and they have all these aggregation dynamics that ensures that the system, is all the time searching, but getting back into the colony state. A gas by definition is something that just escapes, that doesn't take the shape of the container -- it just tries to go always outside. Everything's separated. And I think this is a very important principle for complex organized systems: aggregation. Aggregation is part of the story. Again, of course it's always a conflict between search and in a way getting together when you're thinking swarms. But you need that and that's consistent with the liquid kind of metaphor. But I will be surprised that the gas metaphor applies somewhere.
SFI/Michael Garfield:
Yeah, it's not really going to be doing much thinking if everybody's headed in a different direction, right? So this seems like a good time to bring us to applying all of this thinking. Recently you gave an earlier talk I'll link to in the show notes and you've written a few papers on the ecological crisis and coming up with synthetic biology response to that crisis -- terraforming the earth basically. And I'd love to hear you talk about this and perhaps start by outlining the problem as we see it, if this problem is kind of a euphemism here. You've got this paper you wrote with Simon Levin – “Ecological Complexity in the Biosphere: The Next 30 Years” in Philosophical Transactions of the Royal Society B.
And I think this makes the case for us. Please set this up for us.
Ricard Solé:
Very unfortunately we'll live in this period of time where the decisions we make about the biosphere and about climate are going to be decisive. Mainly because there's something that very often is, I wouldn't say ignored, but not mentioned too often. This is the presence of potential tipping points. Complex systems usually have break points when you push the system in a parameter space -- where you change something slowly but the system responds quickly at some point. Everything points in this direction. Not only global warming itself can create one of these points of no return, but ecological responses to change are very likely to be stresses alike.
This, for example, is something we work on with experts in the field in drylands, which are 40% of ecosystems. These are very stressed systems and they are experiencing changes that might bring some of them into desert states or very degraded states, which, keep in mind, are water keepers. They are really essential for a lot of ecosystem services and more than a third of the people in the world live there. This clearly is something that we have to address. I'm not saying that a technological fix is the solution. I mean, I think that technology has to be part of it, but we need to change a lot of things, right?
So no one who listens to us thinks that this is kind of a magical thing. But the fact that the risk perspective has to be changed because you have the tipping points requires us to start thinking whether or not we have been engineering the planet already. So we have actually modified almost everything, and we are doing everything wrong. It could be a possibility to use synthetic biology, so modifying species and ecosystems. But that's not an arbitrary decision. It's something that we have a whole theoretical framework to think about how to do this in such a way that the engineering we do provides a source of help.
Imagine, for example, in a dryland you put a new species that comes from something that was there and helps retain some more water. This water in nonlinear systems can be amplified and you can help ecosystem to get away from the tipping point. That opens the door for a number of things. We hope that this is some kind of idea that eventually we can actually test in the lab, but it's going to be necessary because we are running out of time and we need to really take seriously the fact that the timing window is shrinking.
SFI/Michael Garfield:
I mean we're finally here now. The paper that I just found immensely interesting is one you wrote with Blai Vidiella in iScience, “Ecological Firewalls for Synthetic Biology.” The conversation I had with Kate Adamala led us right up to the gate of this paper because the question that she seems to always get from people and the question I'm sure you get a lot is, how do we avert a kind of Michael Creighton scenario, an Andromeda Strain or Jurassic Park type of thing. And it's kind of a funny path dependent outcome of the way that chaos and complexity sciences were actually brought to public attention by techno-thriller authors like Creighton who have everyone convinced that the uncertainty inherent in these systems and the inability of people to accurately model all of the possible consequences and like second 10th-order consequences of a new technology make us destined to undermine our own efforts in doing this.
And yet this paper brings a level of nuance to this conversation that I've seen no one else actually bring. So I would love basically for you to lay out how it is that people have been thinking about this traditionally in terms of top-down control and how what you and Vidiella are proposing brings a whole new angle to this conversation. And specifically, what are the kind of mechanisms that you're suggesting here as possibilities for applying synthetic biology for this kind of eco engineering project without releasing horrors into the world?
Ricard Solé:
Well, let me first say that one of the interesting things about the sociology surrounding this is that when people ask, ’how do you know?,’ about what the unintended consequences of this are, I always answer that I can't answer this question because it is non-scientific. Since the seventies, because with the recombinant DNA, there was this idea that whatever we do and release is going to be a threat -- that something really bad can happen. And it was supported by some scientists actually without knowing really if that's the case or not. So it immediately came with the idea that the consequences are going to be bad, you have to put moratoria, and essentially put no research into it.
Nobody knows -- that's the thing that is true. So it's like if you ask me, do you think there is life on another planet, the vast majority of scientists will say, yeah, I think it's reasonable, right? But I believe that -- it's not proof of that. What happens with unintended consequences? The truth is that the few things we know is that if you release something, some genetically modified strain in nature, it dies immediately. It dies immediately because it's in a completely hostile system. And because biodiversity is a firewall to invaders. These invaders that succeed and that we have created -- that's the real threat, like rabbits in Australia or whatever -- are organisms that are large volume masses and usually they are introduced in an environment where you have already degraded everything. Australia's a good example. We should remember that we first killed all the predators, we introduced millions of ships, and in that habitat without predators, the rabbits of course had a lot of opportunities. It was Disneyland. And the thing is the intended consequences, I usually discuss this because we know that, as I was saying, the vast majority of invaders fail to invade ecosystems that are healthy. A few of them succeed but they just remain there.
That's kind of our target -- that you are controlled -- and most importantly the engineering that they want to do has something totally different from the standard engineering. We don't want to predict we will have this ecosystem. We want to provide the source of making biodiversity to keep there, running, because biodiversity is an emergent feature, one that provides resilience to the system – provides, as I said, the mechanism for rejecting invaders. And that is kind of the target. It is a little bit along the lines of David Krakauer’s emergent engineering. That is a philosophically different perspective.
The paper on ecological firewalls and another paper I did with one of my PhD students, Victor Maull, go in this direction. As it happens, in genetic tech engineering, people have been thinking about how do I control a microbe that I introduce into a patient, in the microbe, in the gut? How do I control that? And there's been a lot of work at the beginning like creating suicide kill switches – like, if that happens, the cell will kill itself. As if everything was a problem. When in fact, and that's the point of these papers, in fact, ecological interactions can provide the source of control. You can put something in the system and it's going to be stabilized by the system and remain there -- the biodiversity is not going to be modified. And we have a lot of evidence for that. So you can do engineering and engineering that helps the system -- as I was saying before – by providing the conditions for making things more resilient, and keeping diversity, which is totally essential. But ecology itself provides the firewalls. You are not going to go and do whatever you want because that's not what a rich ecosystem allows to happen. So that's why we think that it's possible to do the engineering,exploiting an emergent property, which is biodiversity, and maybe helping to at least gain some time, that make the systems get far away from tipping points.
SFI/Michael Garfield:
So here are some other areas where it seems like we can see these kinds of dynamics at play. And so, I don't know if you think this is useful or not, but like I think about the shift in thinking between an obsession with antibacterial soaps and then I've heard conversation about the notion of actually like encouraging healthy bacterial communities on human skin so as to prevent invasions. And then at a second level, you're also not encouraging the development of antibiotic resistance. It's funny because maybe this is more of a question for Stefani Crabtree or Jen Dunn, these folks who are thinking about this in terms of archeoecology, but it's common practice for people to pride ourselves in a kind of seemingly increasingly self-loathing way on our ability as a species to have dominated this entire planet, to have invaded every ecosystem.
And yet it strikes me that there's still a kind of a question mark when we look into human prehistory about just how much we were actually responsible for the extinction of prehistoric megafauna and how much our ability to invade the world had to do with instability due to climatic shifts and us sort of capitalizing on the perturbances in these networks. I don't know if you feel like any of that's worth speaking to.
Ricard Solé:
Well, it's complicated question, but the thing is that it's very likely that humans succeeded in part because being equipped with [account??] apparatus that help to start using tools, etc. But the challenge is posed by an environment that is fluctuating -- I don’t know how much happens for that – so I think it's reasonable to think that was kind of the challenge that pushes the brain into getting into more plasticity, in a way, to find solutions. And there was probably a runaway effect, a combination of our upcoming community apparatus and the potential for environment to put us into travel and probably both things were relevant.
SFI/Michael Garfield:
My last question for you, a big one anyway, is about the specific mechanisms that you explore in the model in this ecological firewalls paper. You talk about resource consumer dynamics, mutualism, parasitism, niche construction, and indirect cooperation. I would like to invite you to walk through how you imagine and why you imagine those mechanisms as candidates for preventing these unintended consequences from synthetic biology. Because reading the rigor of this argument put me strangely at ease and I appreciate that and I'd like to spread that ease to listeners as much as possible.
Ricard Solé:
I think one of the objectives of using these different kind of potential scenarios of exploiting standard ecological interactions was to show that you can actually exploit all kinds of things, from competition to cooperation and even interactions that in dynamical terms you read as parasitism. So it's a whole diversity. You aren’t confined to one. Each one of these firewalls, which essentially is a set of ecological interactions that you have to deal with when you put a synthetic organism, have to do with things that we know well. When you exploit a cooperative interaction, cooperators need each other, so one population needs the other. And that can create kind of a lot in effect. You get it under control, you do the function, and your population is stabilized. That all comes just for free, in a way, from the nature of interactions in ecology itself. I know that there are attractors, so why can’t I put myself in one of those attractors, especially if I can remain there, be there when I'm needed, and keep the rest functioning. That's the big key, and it's why we believe that this is something that might actually work.
SFI/Michael Garfield:
You talked about biofilms in this, too. Because we were talking about scaffolding and structure and all of that earlier in the conversation, I'd like to hear you just say a little bit more about specifically how you imagine the biofilm in terms of the way that we can look back into like the very earliest Stromatolites and things like this for ideas about how to handle with environmental toxicity and so on.
Ricard Solé:
Yes, biofilms are very widespread. It's a huge amount of microorganisms that actually live in biofilms, which is a reminder again that this is not multicellularity, but a scaffold that brings the advantages of kind of multicellular organization. And within that kind of system you can actually manage the resources and prepare maybe for other events. We know that some biofilms in different places can actually capture toxins. And it's part of what we know – even, for example, in urban environments, which is one of our targets also for terraforming, because cities can be seen as kind of huge bioreactors.
The point that we have made also is that we could use all the biomass there to actually do something. Why not use something that nobody likes, like sewage, to engineer that, and provide some functionality -- making the cities in a way they are part of the problem, part of the solution.
SFI/Michael Garfield:
And then just lastly -- because I think this was a really, no pun intended, a concrete example of a use case for this kind of stuff -- you talk about the synthetic parasitic firewall and the possibility of engineering organisms that fix CO2 by filling cracks in concrete, which is something I talked about with Kate Adamala, something I've heard Rachel Armstrong talk about as a potential solution to the decay of Venice, Italy. How exactly do you propose that this would work?
Ricard Solé:
The concrete crack for us actually was kind of a revealing thing because it comes from the original starting engineering projects in synthetic biology, this idea that in a concrete crack you have an environment and now it's not very nice for bacteria, but you can find bacteria that live there. Well, this is a huge problem for infrastructures. The idea is if I could bring bacteria there and modify it so that they secrete a mixture of some biomolecule and calcium carbonate -- this has been done, -- they could just live there: they feel a crack, and once they feel the crack, since they are prepared to die, but not for the external world, they just die.
And I always say this kind of a North Korean motif -- that you function and die. But that opens the door for many things. I like to mention Venice because for many years I've been associated with the European Centre for Living Technology there. In the early beginnings of this terraforming idea, we had this project where the Laguna of Venice could have been a target, because you have there all kinds of problems related to toxicity, with [putrification?], and a lot of potential candidates with the species that live there where we can engineer a microorganism to get rid of toxins. And I still think it's a very potential target, but we still need to convince people that this is something that has no travel, because the fear is still there and we need to have a long run yet to convince them that there are good arguments for thinking that this should not be a problem.
SFI/Michael Garfield:
Just in closing it, it, maybe this is a little too speculative, but thinking about all of this stuff again, you know, given the rich seam of literature that looks at networks in not only ecosystems but in the propagation of beliefs and behaviors and economic activity and so on. I can't help but wonder if you have any thoughts in closing on how this work on ecological firewalls might apply to some of the questions that have been raised by people like Joe [??] Coleman, Carl Bergstrom, and others on getting a handle on social science right now as a crisis discipline because ultimately our ability to coordinate for climate solutions depends on us being able to coordinate politically. And yet what we see now are people like Mirta Galesic, Henrik Olsson, and Josh Garland, who have worked on the systemic problems of modern digital communications. Maybe there are ways that we can induce ecological firewalls in our human social systems that allow us to coordinate more effectively?
Ricard Solé:
That's a topic that is not my ongoing research, but I'm really interested and it's been interesting to see that on the one hand you can build a network architecture for misinformation. So you can actually build a network, see how parts interact, see how actually some kind of rich [cloud?] phenomena emerges in the system. But it's a tricky thing because you can build the firewalls in the sense that you can actually try one -- to target the sources of misinformation, target the spreading. But of course we face this problem now because of the echo chamber effect. You could I think identify, know a keystone in this, in this propagation of misinformation, which is in itself is a strange question. It is strange concept, isn't it? It's not information – it’s the deterioration of information. So even if you do that, the problem is that the community that is around is likely to not accept the kind of arguments, the rational arguments that you are bringing. And you know, this is actually I think one of the really fundamental applied problems of complexity, the polarization of society, the role of misinformation, and there's a number of people, like Stephanie Forrest and others that are working on that. We have a huge challenge. I think that whatever is found out in the future will probably need to have kind of a new conceptualization or maybe we need something else that comes to our help.
And that may be artificial intelligence. Something that brings a way of managing this landscape that has this really big conflict, has misinformation. You will say as a rational person that yes, but they have arguments against that. Why is this not working? Because on top of it is the fact that the communities are still polarized. So how do you manage to do that? And I think that artificial intelligence, without being an [apple stall?] of that, but may provide new ways of attacking the problem in a distributed manner, in ways that can break what the [emphasis?] will say -- that the symmetry breaking phenomenon that has to be reversed in a way. So you probably have to find emergent solutions. An emergent solution might not fit into a kind of a political scheme. But I believe that. It's something that I hope is solved because we really need that to be solved.
SFI/Michael Garfield:
New ways of flocking as people. Ricard, this has been awesome. Do you have any final thoughts or questions you want to leave us with?
Ricard Solé:
Well, now that I'm here, I always insist in most of my talks at the end of that, I think that one of the big problems we need to solve in complex systems has to do with things that we have found, what we call universals. I just give you an example because it's deeply connected with the problem of cognition, what's possible, what's not. The fact is that in language, for example, we have all these: Chomsky's grammar -- very formal, well-defined scheme where we can actually fit some universal features of language. But then in the physics and complex networks, we have found all the universals. Scale-free networks in language organization, modularity, all kinds of things that repeat again and again in every language.
So what is the link? What is the connection? Because we are talking about exactly the same kind of entity. And one scheme is very computational going into kind of the deep structure of cognition. The other is more into universal properties that you see as in statistical terms. But we come from the same system and in nature we have that a lot of time. There's the grammar organization, the measures we do that we can interpret in terms of distributions or models or whatever. But we still don't have the link. And finding the link I think will solve a lot of the problems we have, in particular for finding out general loss for complexity.
SFI/Michael Garfield:
Well, that's a call to arms.
Ricard Solé:
Yes.
SFI/Michael Garfield:
Thank you so much for being on the show.
Ricard Solé:
Thanks. Thanks a lot.
SFI/Michael Garfield:
Thank you for listening. Complexity is produced by the Santa Fe Institute, a nonprofit hub for complex system science, located in the high desert of New Mexico. For more information, including transcripts, research links, and educational resources, or to support our science and communication efforts, visit santafe.edu/podcast.