Discussion with Chris Fields and Julian Gough #1

This transcript is automatically generated; we strive for accuracy, but errors in wording or speaker identification may occur. Please verify key details when needed.

[00:00] Julian Gough: The thing I'm working on is a theory.

[00:02] Julian Gough: Cosmological natural selection, which comes out of Lee Smolin's work, which in turn comes out of John Wheeler's idea that big bangs and black holes might be the opposite sides of the same thing. And that you might get inheritance with variation in universes if they reproduce, and that could fine-tune the basic parameters of matter. But that's interesting for what you're working on, because you guys would be very much concerned with information in physics and information in biology, maybe? Would that be fair?

[00:33] Julian Gough: Oh, yeah.

[00:36] Julian Gough: If this theory is true, if your universe is reproduced with Darwinian evolution, the basic parameters of matter end up fine-tuned to produce galaxies and eventually planets and life under the model I'm exploring, because the model I'm exploring, biological life, is actually beneficial for the units of selection that is the universe. So you'd have information already in the basic parameters of matter that's evolved in. So you end up with a lot of very unlikely looking consequences of the relationships of the parameters. I think that would be a really useful extra frame to put around the whole issue. Because you end up with these weirdly unlikely emergent properties and we might be looking for the explanation at the wrong layer. Does that make sense?

[01:54] Julian Gough: Yeah.

[01:55] Michael Levin: Finding unlikely correlations is the sort of thing in biology that drove the search for some kind of evolutionary theory in the first place, right? Because you look at biology at any level and you say, isn't that amazing? This thing adjusts to this other thing exactly how it is. Why is that, right?

[02:13] Julian Gough: I'm going to turn off this background. I didn't realize I had this background on. It's going to drive us all crazy; it's going to flicker in and out. Get rid of the background. How do I get rid of the background?

[02:26] Julian Gough: Adjust background.

Julian Gough: I think the same thing can be said about things. Isn't it strange the way the basic parameters of matter lead to these weird, unlikely consequences in structure formation at different levels in the universe. It's the same explanation, I think, which is evolution. A really good example would be the strong nuclear force. The strong nuclear force is unbelievably attractive, ridiculously comically attractive, until you compress matter to a density that happens when stars collapse at the end of their lives, and then it becomes unbelievably repulsive. But the consequences of that are that the elements that have been formed by fusion in the heart of a star can get distributed up from the bottom of an incredibly deep gravity well back out into the interstellar medium to enrich the hydrogen that's flowing in from the filaments to form the next round of star formation, which will then fuse elements again to build up even more of the periodic table. The star gets to the end of its life; it collapses. The strong nuclear force again reverses at exactly the point where you would like it to reverse in order to distribute those elements back out into the interstellar medium. And that kind of quirk of physics looks to me analogous to the kind of quirks of biology that we ended up using evolution to explain. Isn't it unlikely and interesting that it does this and it has these consequences? So they're the kind of things, and I think a lot of quirks in physics make a lot more sense when they're seen as adaptive mutations, or variations in the parameters that turned out to be good for the reproductive success of the universe and were therefore conserved.

[04:33] Julian Gough: You know.

[04:34] Julian Gough: A beautiful example of that is the fact that there are no nuclei that are stable with five or eight nucleons. What would the mainstream explanation for that be, Chris? If you ask someone why that is the case in mainstream physics, I'm interested in what they would say, or would they think the question's even meaningful?

[05:03] Chris Fields: For white nuclei, the explanation of structure is mainly the shell model.

[05:13] Julian Gough: Sure.

[05:15] Chris Fields: So, four protons and four neutrons is beryllium 8.

[05:24] Julian Gough: Yeah, which is ridiculously unstable, if I remember.

[05:30] Chris Fields: No, it's stable. It's the beryllium we see.

[05:33] Julian Gough: Oh, okay, right, sorry.

[05:34] Julian Gough: Yeah.

[05:36] Chris Fields: Not a shell closure. My shell model thinking is fairly archaic.

[05:48] Julian Gough: Yes. Okay.

[05:49] Chris Fields: But it's an even nucleus.

[05:55] Julian Gough: Okay.

Chris Fields: Lithium-6 would be 3 and 3. I can't recall whether lithium-6 is stable.

[06:10] Julian Gough: Let me have a quick think about this one.

[06:13] Chris Fields: They're all stable for a while.

[06:15] Julian Gough: Yeah.

[06:16] Julian Gough: But the deuterium bottleneck, for instance, just after the Big Bang.

[06:24] Chris Fields: No.

[06:25] Julian Gough: That's an example of where if you had a slightly different stability pattern in the nuclei, you would get runaway messy fusion in the early stages just after the Big Bang, which you don't get because of the deuterium bottleneck. And so the hydrogen is able to get through that density and pressure era where it would otherwise fuse and you are able to build out the complex universe we see. Do you see that deuterium bottleneck as an interesting quirk?

[07:12] Chris Fields: Deuterium is strange because neutrons will decay under almost any circumstances. If you just unbind them a little bit, they're unstable.

[07:28] Julian Gough: Yeah.

[07:29] Chris Fields: So there's no deuterium, though. A weird beast.

[07:36] Julian Gough: Yeah, fair enough.

[07:39] Chris Fields: A friend of mine spent a lot of time trying to find stable configurations of four neutrons and managed to confine them to a very small region of stability space.

[07:55] Julian Gough: I want to go back to beryllium-8 though, because beryllium-8 isn't stable. Beryllium-9 is stable. It's 100% stable. Beryllium-8 lasts for 8 × 10^−17 seconds. So this is what I'm talking about. There's no stable nuclei with 8 or with 5. And even physicists like you will assume that there is. You'll say that beryllium-8 is stable because it's not on our register. It doesn't matter to day-to-day life. But it is the case that beryllium-8 is unbelievably unstable. It lasts for a ridiculously tiny amount of time. The consequences of that are you don't get runaway fusion just after the Big Bang because two heliums can't fuse and a helium and a hydrogen nucleus can't fuse because they would give you a nucleus with either eight or five. So you've got this, and it's really interesting that you thought beryllium-8 was stable. This isn't even something that's that top of mind for physicists. But it has unbelievably powerful consequences for the development of the universe. It's a really interesting quirk because you don't get islands of instability like that until you get up into the radioactive elements again. That's the kind of thing where I think an evolutionary framework makes a lot of sense because the analogy in biology would be something like the way women's hips are angled, not for the woman to be able to walk efficiently, but for the woman to successfully give birth to babies with large heads. You get these islands of instability way down low that have these consequences that are very developmental and look to me like they have an evolutionary explanation, if they have any explanation at all, that evolution is the most likely one.

[10:03] Michael Levin: Could I ask a dumb question: the fact that this beryllium is unstable, presumably that's the sort of thing that varies across the different possible universes and is selected.

[10:21] Julian Gough: Because it would be easier for a beryllia rate to be stable, but that kind of universe would be less reproductively successful.

[10:27] Michael Levin: Right.

Julian Gough: Okay.

Michael Levin: How do you think about the hereditary medium here? What is it that sets, for any given universe, whether these things are going to be one way versus another?

[10:41] Julian Gough: That is a huge open question. We're really exactly the point where Darwin was when he wrote "Origin of the Species," and he was tormented by the fact that he didn't know what the hereditary mechanism was, but he felt there had to be a hereditary mechanism because that explains so much. And that's exactly where we're at. Lee Smolin wrote an interesting paper that tried to give a kind of a matrix explanation for how you could get variation at the bounce where the collapse of mass-energy to a singularity or some Planck length thing bounces to form the singularity of a big bang and that the parameters could vary. There's a matrix approach to how they might vary slightly in that bounce that he's developed, but nobody's paid attention to it and I can't fully follow it either. So it may be flawed, but more likely I'm flawed. Hardly anyone's worked in it. People are not even thinking about this as an area. It's not on people's radar. The thing that is going to vary is the basic parameters of matter. It's going to be stuff like the mass of the electron. It's going to be stuff like the strong nuclear force. But how you get the variation is a wide open question.

[11:58] Julian Gough: Yeah.

Michael Levin: This is super interesting because it gets to one salient point: any genetic material has to survive the bottleneck. In other words, in going from organism to egg to the next organism, what you can't do is obliterate the information. I don't know anything about big bangs or black holes, but whatever we're talking about here has to be stable across that incredible set of transformations. Anything that you have to say about it, I'd love to hear it. The relationship, if any, of that to the thing I've been bugging Chris and everybody else about, which is this notion of platonic space, in the sense that certain aspects of biology — and you could tell me if that's true for physics or not — seem to be set by properties of what we tend to call mathematical objects of different types. What is it that could survive through a refactoring of the universe like that? I wonder if what we're doing, what you have there, is a collapse into some region of the space of what we call mathematical truths before you can bounce out of there. Because it's got to survive that whole thing. I don't know what else.

[13:31] Julian Gough: Mass-energy goes in and has characteristics. Mass-energy comes out and it has characteristics. So the interesting thing is why would the characteristics vary at all, but only very slightly? I don't know what the answer is to that. But you can have a guess at what's going on by looking at how tuned the parameters are, because the basic parameters in our universe look very, very fine-tuned. If you change them very much, everything collapses and you don't get the periodic table, you don't even get atoms. So it's quite fine-tuned. It looks like it's at a local fitness peak. It's way at a local fitness peak.

[14:16] Michael Levin: I have a question about that. How certain are we that if these things were any different we would have nothing versus we are so fixated on matter as we know it that we wouldn't, we can't recognize all the cool stuff that would happen if you did tweak some of these things? Or are we pretty sure that it would literally be nothing?

[14:40] Julian Gough: We're pretty sure that it wouldn't be able to form any kind of structure. Smolin and his team originally varied 8 or 9 different parameters to see how much variation you could get before it lowers the reproductive success rate of the universe by not building out stars and black holes. You lose star formation really quickly if you move many of those parameters by very much at all. There's more work to be done on this. I cannot emphasize enough what an under-explored area it is, but the work that they've done does seem to indicate definitely that this is a remarkably fine-tuned universe. There's not a lot of give in a lot of those parameters at this point.

[15:27] Chris Fields: I'll just make a couple of remarks. One, there's a huge literature in this area that I'm unfamiliar with, but if you look on arXiv and put in "bouncing universe," you'll probably get dozens of papers per month for the last 10 years.

[15:43] Julian Gough: But they all have completely different theories. The problem is the bounce is exactly where general relativity breaks down and quantum mechanics breaks down. It's a beautiful, fun place to play if you're a theoretical physicist, because you can do anything. A huge number of papers have come out on the bounce, but there isn't coherence in them because they've got too many free parameters at that point. I don't even think there is a consensus. What would you take the consensus view to be on the bounce, if there is such a bounce? What's your general feeling for the state of the field at the moment and what they think is happening?

[16:30] Chris Fields: Yeah, I lost you all together there.

[16:32] Julian Gough: I'm sorry, Chris. I was wondering, what's your sense of the general state of opinion in the field about black hole–big-bang bouncers?

[16:45] Chris Fields: Oh.

[16:46] Julian Gough: Because my sense is it's a very chaotic area where people are very free to speculate on it. There isn't really a consensus opinion.

[16:53] Chris Fields: Two things. One, I'm not familiar with this literature. It does seem to vary all over the place. Discussions of fine tuning go back way before this bouncing universe idea.

[17:16] Julian Gough: Sure.

[17:18] Chris Fields: With anthropic arguments and all that.

[17:21] Julian Gough: Sure, yeah.

[17:23] Chris Fields: Yes, in answer to Mike's question, too much variation in even just values of h-bar and the electron charge will lead to unstable nuclei.

[17:40] Julian Gough: It's not like they lead to a different stable equilibrium. They don't; they just lead to mush.

[17:49] Chris Fields: In the quantum gravity regime, all bets are off on semi-classical thinking, which is the whole black hole-as-collapse idea. And the idea of a singularity completely goes away.

[18:27] Julian Gough: Yeah.

[18:29] Chris Fields: I would say the nearly dominant way of thinking is that the whole space-time structure is emergent.

[18:50] Julian Gough: Right.

[18:51] Chris Fields: Which calls semi-classical thinking in any of these critical regimes really deeply into question.

[19:03] Julian Gough: Yeah.

Chris Fields: But I would back up from that and ask a variation on Mike's question, which is: natural selection always happens in some environment. So the other question besides the one about stability of information is what environment is the information stable in? And again, Mike, in your language, this may be some platonic space, as an environment. But then one has to come up with a notion of how that environment works and what it means for that environment to implement a fitness function.

[20:03] Julian Gough: Yeah.

[20:04] Chris Fields: That's the other kind of immediate background question from my point of view.

[20:10] Julian Gough: Everything we observe is natural; selection is always operating inside an environment. The trouble is the environment is the universe. One way of thinking about it is that each is a separate bubble of space-time. Each is its own; it's both organism and environment. It has to bring all its energy with it for its lifetime. Our universe: so much of what our universe is doing is the frugal transmission of energy over vast spans of time. If you look at a star, the energy output from a star is extremely frugal and extremely long. It's nothing like a random process. It's nothing like a forest fire. It plateaus for billions and billions of years. In the case of red dwarf stars, for billions of years. So you can see if our universe evolved, it looks like it's both environment and organism in its development and its life cycle. It's not getting extra energy from outside. It has to bring it all with it. Which means most of what it's doing is producing and directing energy to places that it can use that energy to build up order. And that happens at increasing scales of order in decreasing areas of space, because it has to. It can't have massive order over the entire thing, because there isn't enough energy to do that. What you see is the periodic table being built out by rounds of star formation and distributed. There are planets and moons, which are much, much smaller than stars. There are surfaces there where chemical activity can rapidly iterate evolutionary algorithms and build out biological life. Biological life, once it gets to a certain level of complexity, can build out technological life even faster, because that's directed evolution. It's extremely fast. There's a lot of directed energy flow to do very constrained, rapid complexification and self-complexification and self-ordering. It looks analogous to metabolic pathways in biological life. The energy that's flowing through the system is organizing the system. The big argument I was making before the James Webb came online was that we're going to see the energy flowing through the system organizing the structure formation of the universe right from the start. It's not going to be slow, random, arbitrary, bottom-up structural formation just passively happening through gravity alone. You're going to get supermassive black holes by direct collapse. They're going to send out jets. The jets are going to do a lot of structural work on the early universe that will play out in the later structures we see, the voids, the filaments, and so on. They're going to build out galaxies rapidly and efficiently. That would be one answer to the question: a universe, just by the nature of what a universe is, is both organism and environment. The other question then is, once a bubble of space-time bugs off from a parent universe, forms a child universe, is there a shared environment for those bubbles of space-time? Is there a shared environment outside our universe for universes? I don't know the answer to that question. I'm doing a naive theory that assumes that the parent universe and the child universe separate and can't affect each other consequentially, that there is no energy coming in from the outside, that the universe is genuinely separated from the parent. Because if you don't go with that simple model, it gets into such speculative territory that you can't really play around very much.

[24:21] Chris Fields: That's the kind of standard classical multiverse view, and this was an early view of Tegmark. I don't know whether he still holds it or whether he abandoned it long ago, but the kind of classical multiverse view is that every possible combination of parameters exists somewhere.

[24:50] Julian Gough: Yeah.

[24:51] Chris Fields: Implicit in that is they all have the same kind of space-time structure.

[25:00] Julian Gough: Yeah.

[25:02] Chris Fields: I see no reason in the world to make that latter assumption if you're varying everything else.

[25:12] Julian Gough: Everett is an odd one. Everett is taking the rules of the basic parameters of matter in our universe, assuming they are the rules and they don't vary. Then he's running every possible variation on how that could play out. But he's not varying the basic parameters of space-time.

[25:28] Chris Fields: That's the difference between the classical multiverse view and the Everett interpretation of quantum theory, which is totally different.

[25:36] Julian Gough: Yeah, totally different, yeah, But.

[25:37] Chris Fields: If you were to take Everett back into the early inflationary period, you get a combination of the two.

[25:48] Julian Gough: You would get a combination of two, but you end up with this absurd kind of infinity of infinities of universes, which is fine, but it doesn't, I don't think it's nearly as satisfactory as an evolutionary model where you have an evolutionary tree of universes where those branches that are more reproductively successful and produce more black holes pass those traits on to child universes, which then are more likely to produce more black holes. Some of their offspring will produce more black holes, some less. The ones that produce more have more offspring, and you've just got a very straightforward evolutionary movement towards greater reproductive success and towards more black hole production. The model really seems to have a good fit with what we're seeing in our universe. The three-stage model that I'm proposing and exploring, and the one that made predictions that the James Webb has verified, makes me feel much more confident in pushing this theory at you, Chris. There are three modes of black hole production. One must be the primal way that the earliest, most primitive universe is reproduced, which has to be just by direct collapse. A small number of very large offspring produced just by the direct collapse of some kind of proto-matter. Imagine something simpler than hydrogen. No big periodic table, because the periodic table is a later evolutionary breakthrough that requires a lot of complex structure to have evolved to build it out. These direct collapse supermassive black holes are the earliest form of reproduction. The supermassive black holes we see in our universe, the ones with masses of millions of times the mass of our sun, or billions of times, would have formed by direct collapse, and they have to have done it extremely early when the universe was extremely smooth, as we see from cosmic microwave background radiation, because they couldn't do it later. Once you've got a lot of density variations in the gas, any large scale collapse will nucleate out into star formation. You have to have an early wave. This is an evolutionary prediction. You have to have an early wave of direct collapse supermassive black holes happening within the first 150 million years after the Big Bang, when the gas is still smooth, preceding star formation, and then they drive galaxy formation and star formation. That's what the James Webb is showing us.

[28:19] Julian Gough: The James Webb is showing us galaxies that are forming rapidly and early and around supermassive black holes that are larger and larger as a proportion of the galaxy the further back you go, which is exactly what this model predicts. That's very suggestive. Stellar collapse black holes are a later evolutionary breakthrough. It's more frugal use of the matter and energy in a universe to build out a larger number of smaller black holes, which can produce full-size universes, because positive mass-energy and negative gravitational energy net out to 0, so you can build a full-size universe even if the parent black hole gets smaller. Then you have the breakthrough to small technologically produced black holes, which would be what we and other life forms like us ultimately would all converge on, because that's the most effective form of energy production in our universe, is just dropping matter into small black holes. If you can create small black holes as energy sources, you can get up to 42% of your mass back as energy from any material. That would be another breakthrough. Having those three forms of reproduction, each requiring more of the periodic table, each requiring more complexity, each building on the previous breakthroughs, looks like an attractive model worth exploring, worth getting more minds involved in. It has consequences for your work, because if that's the case, the matter itself has been fine-tuned. There's a kind of co-evolution going on here where universes need to maintain all three forms of reproduction. You can't change the periodic table in a way that stops direct collapse supermassive black holes forming because they're needed to make the galaxies and the stars and so on. You can't break stellar formation because they're needed to make the periodic table and planets. But within those constraints, you can fine-tune, especially as you get further up the periodic table; you can fine-tune phosphorus, you can fine-tune calcium. And so biological life should be relatively easy to come about in a highly evolved universe where that benefits the reproductive success of the unit of selection that is the universe. So you should be getting very unusual and unlikely emergent properties of matter that are life-friendly in such a universe. I do think that's what we see.

[30:51] Julian Gough: I do think that's what we see.

[30:54] Julian Gough: I think it's such an under-explored framework for thinking about all this.

[30:58] Michael Levin: Another question, let's see if there's a parallel to this. One of the things we focus on in our lab is the plasticity of what happens with a fixed genetic medium. So in other words, you have a genome of a particular kind and under default circumstances, you get a frog, but you can also get xenobots and they'll do all kinds of cool things that they don't do. I always talk about these different examples where the exact same hardware can give other different outcomes. Is there any sense of this plasticity? Now, given Chris's question before about what's the environment, I don't know where the differences would come from, but I'm curious, in any given scenario, whatever the hereditary medium is here, how much variability do you think that allows to what happens after that? Or is it really one-to-one, though?

[31:53] Julian Gough: But I think the plasticity you're talking about is a downstream consequence of evolution at the level of universes. I think the periodic table is extremely life-friendly. And you can, you're exploring that territory by doing the Zeno stuff. Life isn't a fragile, arbitrary accident that happened once and could fall apart at any second. It's a very likely thing to happen once the conditions that evolution has tuned it for are in place. And I think you're seeing examples of that with some of your work. You're seeing, "holy crap, matter wants to be alive" — that's a bit of a t-shirt slogan way of putting it, but there's truth in that. Matter wants to be alive. That's why a lot of it's here. I did do a very t-shirt version of this, which I mentioned to you, Michael, and to you, Chris: biology is the first half of the periodic table coming to life, and technology is the second half of the periodic table coming to life. The whole table is tuned for a developmental process that leads to exactly what we're doing here: intelligent life forms discussing the future and what we're going to do next through technological tools that will build other technological tools that will allow us to explore the possibility space for this particular universe and its future through life and technology intertwined. We're part of a developmental process. We're not an accident at the edge of something.

[33:39] Michael Levin: And so given that, what's your preferred take on the paradox of where is everybody?

[33:46] Julian Gough: That is a really interesting question because under this model you should have life all over this universe. I don't have a clean answer. I have several speculative answers. It doesn't look to me like this universe is optimized for life on the surface of rocky planets in a narrow habitable zone. We've got four rocky planets. Only one of them has been able to hang on to water. We're a really nice example perhaps of life—matter really wants to be alive and it will do it even if the circumstances are not ideal. If you look, where's the liquid water? The liquid water is mostly under the surface of icy moons. That doesn't have a Goldilocks zone. Everywhere has a habitable zone if you're an icy moon going around Saturn or Jupiter or any large gassy planet, because what's keeping your liquid water ocean liquid isn't the heat from something eight light minutes away, which is a very inefficient way to boil water. It's the tidal friction from the eccentric orbit around the gas giant. That tidal friction transmits energy very directly to exactly where it's needed. The analogy I use is icy moons keep their water liquid by an electric-kettle method where all the energy goes straight to keeping the water liquid, whereas on the surface of a rocky planet you're keeping it liquid by setting fire to a warehouse eight miles away. If this probably isn't our optimum, maybe chemistry takes longer on icy moons and that's where most life forms are developing. Most of the stars are actually red dwarfs. We just found out in the last week or so that the TRAPPIST system, which is a red dwarf system, seems to have quite a rich atmosphere on at least one of its planets. We've been able to observe one of the planets close enough to get atmospheric spectroscopic readings, and it has nitrogen; there's a lot of interesting stuff in the atmosphere. So, given that red dwarf stars are far more numerous than our G-types, maybe that's where life is optimized. Maybe it just takes longer, and we're early.

[36:32] Julian Gough: I don't know.

[36:33] Julian Gough: It's an excellent question. I would predict if this model is correct that we will start seeing signifiers of life in the atmospheres when we get close enough to see them on multiple worlds. Life should be very common. Why isn't it operating at a scale that we can see? Why isn't it visiting us? That's a whole other question. It's significant that stars are so ridiculously far apart. If this is an evolutionary universe, there's a reason why the stars are so far apart, because they're a light year apart — it's a very difficult distance to cross; it makes it very difficult to get outside of your solar system. You can think of solar systems as cells and the membrane is the awful space to the next solar system, so they're essentially isolated in certain ways. They form structures. A galaxy is an autopoietic structure. A galaxy maintains its shape, maintains its structure. The stars all get replaced; some last a very long time, but there's a lot of star production and star death, and the spiral galaxy goes on and on. It's very autopoietic. If you think of the solar systems as the cells of it, they are separated from each other. You can't get runaway Petri-dish life forms that hop across the galaxy. It's very difficult for them. It seems to be of an evolutionary benefit if this model is true that each life form gets time to explore its own possibility space, its own solar system. I don't know why; maybe that is a more efficient way of reproducing. Maybe it gets more small black holes out of that if everybody gets to explore their own solar system for a while.

[38:41] Julian Gough: I don't know.

[38:42] Julian Gough: It's a good question. It's a tricky question, Michael, because there should be life all over the universe, and I think we will see signs of it. I'm very interested in the Europa Clipper, because that's going to the icy moon Europa to try and find out what they can. It was very interesting that on Enceladus, when we were able to measure the water that came from the interior out through the ice volcanoes — it's such a small moon that it squeezes water out — it contained loads of phosphorus, which they had not thought it would contain. It seems to have the nutrients you would need for life, which is interesting.

[39:19] Chris Fields: From a straight biology point of view, you'd expect most life in the universe to be microbial, with some structure or other.

[39:30] Julian Gough: Microbial should be.

[39:31] Chris Fields: Maybe not our kind of microbial life, but no reason. I really do expect, say, DNA, but something like DNA, perhaps?

[39:40] Julian Gough: Something like it. I think one of the evolutionary frames, again, if you look at the chemistry of life through an evolutionary frame, phosphorus looks very fine-tuned. Phosphorus — I would imagine something like ATP is going to be very common to life forms in very different worlds, because it's hard to see what's going to do a better job than ATP. And it's hard to see what's going to do a better job as a medium than liquid water. You can think of liquid water as fine-tuned by evolution because it has very unusual properties. If water just shrank instead of expanding when it freezes, like most bloody liquids, you wouldn't be able to get life; oceans would freeze from the bottom down and life would get killed every winter. So icy moons wouldn't be able to hold liquid water oceans, which is also suggestive but doesn't prove anything. So water looks fine-tuned. So I think this has really interesting implications that are worth exploring, Amy.

[40:53] Michael Levin: I've been thinking about this water thing for a few minutes already. Do you guys know, because I seem to recall the deal is that the hydrogens are at a particular angle, which is why the crystals don't sink, right? What's actually responsible for that?

[41:17] Julian Gough: I'm not good enough to give you a detailed one on that. It is something to do with the hydrogen bond angles. They are at an angle where, strangely enough, the consequence is very good for life.

[41:31] Michael Levin: I was curious what sets those angles? If we ask why the angle is this and not that, what's under there?

[41:39] Julian Gough: Again, what you're talking about is relationships here. It's like the relationship of hydrogen to oxygen. You end up with a really nice result where the hydrogen bond angles when it's liquid give you a more spacey water, and then when it locks into a frozen form the bond angles are adjusted to and it just ends up smaller, and it's not what normally happens with most liquids. It's about the relationship between hydrogen and oxygen. You can see how evolution could fine-tune that. There would be other universes where those bond angles are different, where water expands when it turns to ice, and they won't be as reproductively successful, so there will be fewer of those universes. We're likely to be in a more successful one. But that's the kind of thing I'm talking about. You can actually look for evolutionary explanations for these quirks of physics that have such powerful consequences. A lot of what's happening is relationships. It's not just that the parameter is set by evolution and that's where the information is stored. That parameter relative to this parameter will lead to these consequences. There's a lot of complex emergent consequentiality that comes from the relationships. Even though you've only got 27, 28, maybe 30 basic parameters, you've got a huge amount of information stored in their relationships to each other. How that's been fine-tuned?

[43:22] Julian Gough: You know, it's...

[43:25] Chris Fields: I think that a chemistry in which the nuclear physics was different enough that you would get different bond angles and water would be a chemistry that didn't have water.

[43:45] Julian Gough: Yeah, exactly.

[43:48] Chris Fields: Or anything analogous to it.

[43:50] Julian Gough: Yeah, Well, an interesting thing.

[43:53] Julian Gough: Go on.

[43:54] Michael Levin: That makes sense, but I'm curious, and maybe I need to look this up. Given whatever the parameters in our universe are, what's the genetics explanation I'm looking for, not the evolutionary scale? What is it that sets it? Is there anything that we know about the universal constants that would allow us to derive the angle? Do we know what it comes from? Could we have guessed it if we didn't measure it from other stuff we know?

[44:27] Chris Fields: I expect good physical chemists would say this is how it works. I don't know physical chemistry.

[44:34] Julian Gough: I think this is something that comes under more-is-different, Anderson's favor, that I think that's a consequence that emerges at the chemistry level. It's not obviously absent from the physics level, but there are a lot of open questions and a lot of interesting explorations that could be made here. It is the case that water has very unlikely properties that are a result of its chemical makeup and they lead to life. If life leads to small black holes, as I think it does, then it's good for the units of selection that is the universe. Weird quirks of physics — if you want to explain them, you look for the downstream consequences in the reproductive success of the universe. That keeps happening. A lot of weird quirks disproportionately often seem to lead to downstream consequences that would be good for the reproductive success of the universe, which is what you would expect in an evolutionary model where you can think of those conserved quirks as analogous to mutations that are beneficial and that are going to be conserved as a result. I also mentioned before, Michael, the idea of the CNO cycle. Did I mention that one to you?

[46:13] Julian Gough: I don't think so.

[46:14] Julian Gough: Let me throw this at you. This is an exaptation argument for chemists, how chemistry would have evolved. I think this is suggestive. It's not proof of anything. It's very suggestive that fusion in stars has two mechanisms. There's the proton-proton chain reaction. Look at that through an evolutionary frame. That has to be the original fusion mechanism in stars, because it just requires protons. It doesn't require more complex structures. So the proton-proton chain reaction would have first evolved so that stars could be stars and could fuse matter and do their thing. And then you get stellar collapse black holes, and you can see how that can help universes reproduce. But then you've got the CNO cycle. The CNO cycle is the carbon, nitrogen, oxygen cycle, which acts as a kind of catalytic reaction that allows for a lot of fusion in stars. It's more energy efficient. You can build out more stars from less matter. You can therefore get more reproductive success because you can get more stellar collapse black holes. So CNO would be an evolutionary breakthrough that happened to help star fusion; it would have been conserved because it helps star fusion and stellar collapse black hole production. Carbon, nitrogen, and oxygen would have been on top of the proton-proton chain, which was the original fusion mechanism. Evolution would have had to have stars in the first place, because you're building carbon, nitrogen, and oxygen with stars. That can't be the primal fusion mechanism because it requires stars to have formed first. So CNO evolved later in the history of universes. And then what is life built on top of? It's built on top of carbon, nitrogen, and oxygen, which in this model would have evolved originally for fusion but would then be accepted for biological life because they're already there. The breakthrough to biological life is such an unlikely thing in the first place. You can't have it just happen from scratch. It has to build on things that were developed for completely different purposes. It's classic exaptation. The CNO cycle, great for fusion, turns out to be adapted for the development of biological life, which leads to the greater reproductive success of universes. That is a really suggestive and interesting thing.

[48:56] Michael Levin: Yeah, for sure.

[48:57] Julian Gough: It's really suggestive. It also has an interesting logic to it, because you would have to say that the periodic table essentially evolved. There was an era where there was just something like hydrogen. Then there was a later era where there was something like hydrogen and helium and they were making little stars and that. And there was a later era where there was hydrogen, helium and carbon and maybe nitrogen and oxygen. Then there was a later era where you get throwing some calcium and some sulfur and now you've got biological life and you're exploring the possibility space over many generations of universe. There's a beautiful logic to it. And the raggedness at the end of the periodic table is exactly what you'd expect. In an evolved periodic table it's evolved that far and it's ragged at the ends. It doesn't matter so much that it's ragged at the ends. It used to be ragged further down the table.

[49:50] Michael Levin: Another aspect to wonder about is the size of the genome. You could imagine we have 4 forces and a bunch of others. Now the question.

[50:09] Julian Gough: They keep arguing is the weak force part of the electromagnetic force?

[50:13] Michael Levin: But the bandwidth — what I'm talking about is the informational bandwidth for the early parameters. We have some number of them, but as we know in biology there are really stripped-down genomes and massive genomes with lots of bandwidth for all kinds of junk. What would you guess around that? Do different universes across this variation you're talking about have widely different numbers of fundamental forces or settings?

[50:44] Julian Gough: You can imagine evolution going down totally different paths that are nothing like our own. They're very unknowable. You can imagine a very different genome. I think evolution, because of the crazy bow tie that they're stuck with, because it's the black hole–big bang is how the earliest universe is reproduced. They have to conserve that. You tend to conserve whatever early means of reproduction you had. We still fertilize eggs in salty water as our means of reproduction because our jawless fish ancestors did it. You get stuck with it. So they have to do that. Then it's a matter of what can safely and efficiently pass through that bow tie, that compression point. It may not be a huge complex genome. It's just harder to get it safely through the compression point without variations messing up the whole universe. If you've got a small but sufficient number of basic parameters that you can vary one or two of very slightly, then you've got a stable reproductive strategy where you can explore the possibility space. What's happening is you're getting emergent second-order information structures that act as a kind of supplementary genome. So if you look at the periodic table, that's 96 elements with all kinds of characteristics, but it's built out of these sub-units, protons, neutrons, and electrons. The periodic table has all these emergent, essentially gene-like properties that have enormous developmental consequences for the universe. I would say it's a mistake to look for the entire genome in the basic parameters of matter. What you're looking for is a simple genome in the basic parameters of matter that's capable of building out essentially supplementary more complex genomes, which then build out and so on. Do you see what I mean?

[53:01] Michael Levin: I certainly agree. This is one of the things we study extensively: the non-mapping between the actual details of the genome and the results it's capable of, right? That's—I'm totally on board with that. We're not, which is why wheat and things like this have tons more DNA than we do. I completely get it. That space between the hardware specification and the final outcome is massive and it has all kinds of problem-solving competencies of its own. But nevertheless, we do, at least on the biology side. You have operations, right? You have to think about the variation operator. We have mutations that will change values of existing slots in the hereditary material. We also have other operators that will extend the whole thing or cut it or reshuffle it. I know we don't really know any of these things for your kids, but if we were going to make—what I'm imagining is to flesh out this analogy, to really make a list of the kinds of things, the fundamental, not the details, but the fundamental things we see on the biology end, so that at some point in the future, people who are developing this theory can try to fill in: we're going to need to say something about how this is handled and how that's handled. Maybe some things won't map for sure, but it seems, at least in biology, for all the things that can happen without any changes in DNA, the fact is, we do have a bunch of operators that not only tweak the alleles, but actually change the bandwidth of the bow tie.

[54:46] Julian Gough: That's very interesting. Definitely. Beautiful area to explore. Sorry, Chris, go.

[54:51] Chris Fields: I was going to say to go back to the discussion very early on.

[54:56] Julian Gough: Yeah.

[55:00] Chris Fields: One can think about black holes in multiple ways.

[55:04] Julian Gough: Yes.

[55:05] Chris Fields: Well, one way is just a horizon.

[55:07] Julian Gough: Yeah.

[55:10] Chris Fields: Which for a big black hole can encode lots of information.

[55:20] Julian Gough: Yeah, yeah, yeah.

[55:21] Chris Fields: Not just fundamental parameters, but lots of information.

[55:24] Julian Gough: On the horizon, yeah, sure, yeah.

[55:27] Chris Fields: Or one can think of it in this classical and semi-classical idea of a singularity. And we know that idea breaks down at some point.

[55:49] Julian Gough: Consequence of the, obviously, over that realm.

[55:54] Chris Fields: The collapse story makes you want to think in singularity terms. Then the bandwidth is very small.

[56:04] Julian Gough: Yeah.

[56:05] Chris Fields: But the geometry carries an enormous amount of information. That is not going through the interface, because the geometry itself becomes the environment. And that geometry must encode all of the extra information.

[56:33] Julian Gough: Yes.

[56:37] Chris Fields: Without this kind of classical collapse picture, a black hole is just an interface, a boundary. It looks like any other kind of system. It's just an interface that has some informational bandwidth.

[57:05] Julian Gough: Yeah.

Julian Gough: Does this have implications for an evolutionary theory, positive or negative? Do you see this as good or bad for an evolutionary approach?

[57:14] Chris Fields: I think the fundamental question for an evolutionary approach is what's the environment? Where is the additional information that does the sorts of things that Mike was just talking about? And I don't think you can get away with it without answering that question.

[57:39] Julian Gough: Yeah.

[57:40] Chris Fields: And you can try to build it into space-time, but then you have to assume space-time, which is a very fragile assumption. I strongly suspect that. 20 years from now, we'll think of space-time as just an emergent phenomenon.

[58:00] Julian Gough: It's quite possible. We can still get a lot of value from just exploring the idea of a universe as an evolved membrane-bound kind of organism that is both organism and environment. We can go on to explore what the meta environment for universes is later on. But there's still a lot of value in unpacking the consequences inside our universe of an evolutionary history that fine-tunes the basic parameters. I still think we can do a lot of work with that in the absence of a clear theory of what's the meta environment for universes as a whole.

[58:46] Julian Gough: If you see what I mean.

[58:51] Chris Fields: I think the fundamental question is, what is the relationship between the standard model parameters, which are defined in terms of space-time, and space-time as a structure?

[59:14] Julian Gough: Right, yeah.

[59:16] Chris Fields: And that question isn't answered. But I guess I would characterize the question "how does space-time relate to everything else" as, in a sense, the question of quantum gravity.

[59:42] Julian Gough: Yeah.

[59:45] Chris Fields: But I think assuming a space-time structure allows you to assume almost anything.

[1:00:01] Julian Gough: Maybe.

[1:00:04] Julian Gough: I don't think an evolutionary model generates extra problems that weren't there already. I think it actually solves some problems and has good explanatory power and some predictive power because it is actually predicting the early-universe structure formation we're seeing, which is a very suggestive positive for the theory, for the approach.

[1:00:25] Julian Gough: Yeah.

[1:00:27] Julian Gough: This has been a very enjoyable and interesting conversation. I hope you found it. Do you find it of interest? Are you intrigued by this, or do you think it's almost certainly wrong? What's your take on an evolutionary approach?

[1:00:46] Chris Fields: I see it through the lens of the questions that I just asked.

[1:00:49] Julian Gough: We're trapped inside our end of one. We're trapped inside our single universe that we have access to. The bigger questions are extremely important, but if we can't answer them, there are other questions that we can explore and answer in the meantime that are going to be fruitful. Those larger questions at some point need to be answered. But we can still do valuable work in the meantime inside our N of 1. If you were stuck inside an elephant and you didn't know anything about the entire other biological realm, you just had this one elephant that you were stuck inside trying to understand, you can still understand a lot about the elephant by studying the inside of the elephant and through an evolutionary lens and going, why is this, what is this? How, it's got salty blood. That's interesting. Maybe it evolves in a salty water environment. It's got huge air sacs inside it. That's interesting. You can learn a lot about it just from studying it through this lens. And that's true for our universe too.

[1:02:02] Chris Fields: That is the task of evolutionary cosmology.

[1:02:09] Julian Gough: Is there anyone you think I should talk to about this? Do you know anyone who's interested in this field who was interested a few years ago and then put it aside because nothing was happening? Things are happening now and I want to bring more people in.

[1:02:22] Chris Fields: I don't know the bouncing universe literature. I just know that it's big.

[1:02:30] Julian Gough: Yes.

[1:02:32] Chris Fields: Obviously that's the place to start.

[1:02:36] Julian Gough: Yeah.

[1:02:38] Chris Fields: And to understand what's being done there. I'm not an astrophysicist. I don't know the detailed physical chemistry. I did nuclear physics long ago. So I know about the raggedness of the periodic.

[1:02:59] Julian Gough: Yeah, it's pretty ragged.

[1:03:01] Chris Fields: It's ragged all along the line of stability.

[1:03:04] Julian Gough: It's fascinating; it looks evolved to me. If you look at the periodic table through an evolutionary lens, it's got a lot in common with DNA. It looks evolved. It's messy, it's got workarounds, but it kind of hangs together. Just looked at in its entirety, the periodic table is fascinating. And if it did evolve, there's an evolutionary history in that periodic table. Hydrogen came first. As you go up the periodic table, they evolved more recently and tend to be more unstable and more fragile. You're moving from the parts of the periodic table that were necessary just for the direct collapse of mass black holes up the periodic table. Now you need them to make stars. Now you need them to make life. Now you need them to make technology. It's very, very interesting looking at it through an evolutionary lens.

[1:04:06] Julian Gough: Yeah,

Discussion with Chris Fields and Julian Gough #1

Watch Episode Here

Listen to Episode Here

Show Notes

Transcript

Related episodes

“The Unreasonable Effectiveness of the Behavioral Sciences in Developmental Biology and Biomedicine”

Discussion #1 with Katrina Schleisman and David Burke

Conversation with David Resnik 1

Discussion with Chris Fields and Julian Gough #1

Watch Episode Here

Listen to Episode Here

Show Notes

Transcript

Related episodes

“The Unreasonable Effectiveness of the Behavioral Sciences in Developmental Biology and Biomedicine”

Discussion #1 with Katrina Schleisman and David Burke

Conversation with David Resnik 1

Thoughtforms Life Podcast