#433 – Sara Walker: Physics of Life, Time, Complexity, and Aliens

0:00:00 The following is a conversation with Sarah Walker, her third time in this podcast.
0:00:05 She is an astrobiologist and theoretical physicist interested in the origin of life and in discovering
0:00:12 alien life on other worlds.
0:00:15 She has written an amazing new upcoming book titled Life As No One Knows It, The Physics
0:00:21 of Life’s Emergence.
0:00:22 This book is coming out on August 6th, so please go pre-order it now.
0:00:29 It will blow your mind.
0:00:32 And now, a quick few second mention of each sponsor.
0:00:35 Check them out in the description.
0:00:36 It’s the best way to support this podcast.
0:00:39 We got Notion for Notes, Motific for LLM Deployment, Shopify for E-commerce, Better Help for Mental
0:00:47 Health and AG1 for delicious, delicious multivitamin drink.
0:00:52 Choose wisely, my friends.
0:00:53 Also, if you want to get in touch with me or to work with our amazing team, go to lexfreedom.com.
0:00:59 And now onto the full ad reads.
0:01:02 No ads in the middle.
0:01:03 I try to make these interesting, but if you skip them, please still check out the sponsors.
0:01:08 I enjoy their stuff.
0:01:09 Maybe you will too.
0:01:11 This episode is brought to you by Notion, a note taking and team collaboration tool.
0:01:16 I’ve been using it recently for note taking on academic papers, specifically machine learning
0:01:22 papers.
0:01:23 And there’s a lot of machine learning papers.
0:01:26 And it’s straight up just a great note taking tool.
0:01:30 But beyond that, it’s a great collaboration tool for the note taking process and the whole
0:01:35 project management life cycle.
0:01:37 So it combines note taking, Wiki’s project management, and then there’s an AI assistant
0:01:42 that can summarize everything and anything.
0:01:44 And you could ask questions across all of those things.
0:01:48 So it’s not just for a single document across all the documents.
0:01:52 And obviously people are collaborating on those documents so you can ask questions.
0:01:55 What did this person do?
0:01:56 What is the status of this project?
0:01:59 So on, so forth.
0:02:00 It’s just a really nice integration of LLMs.
0:02:03 This is the fundamental question with LLMs.
0:02:06 How do you leverage the obvious power that they possess to be useful to whatever tasks
0:02:13 that we do?
0:02:14 Like what is the actual product here?
0:02:16 And so Notion leverages them extremely well where the product is team collaboration on
0:02:23 notes, Wiki’s and project management.
0:02:25 So really well done.
0:02:27 Love to support people that do a great job of building a great software product.
0:02:32 Try Notion AI for free when you go to Notion.com/Lex, that’s all lowercase Notion.com/Lex to try the
0:02:39 power of Notion AI today.
0:02:41 This episode is brought to you by a new sponsor, Motific.
0:02:45 It’s a SaaS platform that helps businesses deploy LLMs and in general generative AI that
0:02:52 are customized with RAG, Retrieval Augmented Generation, on organizational data sources.
0:02:59 Obviously, these kinds of data sources are often super sensitive and that’s where Motific
0:03:04 comes in.
0:03:05 They help companies with security and compliance.
0:03:08 A little background here, since I had to do the deep dive myself a while back, Motific
0:03:12 is created by Cisco, specifically Cisco’s Outshift Group.
0:03:17 And Outshift is doing cutting edge R&D stuff in Cisco.
0:03:22 So Cisco has very, very, very long track record and reputation of working with giant businesses
0:03:29 and helping them out in not messing stuff up.
0:03:33 When you’re dealing with sensitive data and when you’re dealing with businesses that make
0:03:37 a lot of money and already have products that bring in a lot of money and a lot of people
0:03:43 rely on, you don’t want to mess stuff up.
0:03:46 I think specifically this task of taking organizational data that’s private to the company that has
0:03:55 to remain very secure, very sensitive data, using the power of LLMs and search via the
0:04:03 RAG framework on that data is super, super powerful.
0:04:11 I think companies that do this well and quickly, which is what Motific helps with, will win.
0:04:19 Because the productivity gains, nobody knows, but I don’t think there’s a ceiling.
0:04:25 So it pays us off to play with LLMs, but do so in a secure way.
0:04:32 Visit MotificAI to learn more.
0:04:34 That’s M-O-T-I-F-I-C dot A-I.
0:04:39 This episode is also brought to you by Shopify, a platform designed for anyone to sell anywhere
0:04:45 with a great looking online store.
0:04:46 I have a site set up, blackstreamer.com/store, has a few shirts on it, took a few minutes
0:04:51 to set up.
0:04:52 Super easy.
0:04:53 They integrate third-party apps.
0:04:55 I did that for the on-demand printing, so I don’t have to think about any of that.
0:04:59 All you do is upload the design, the shirts are sold and shipped.
0:05:04 It’s kind of an interesting experiment for me to understand how people look at you when
0:05:09 you have a t-shirt with nothing on it and a t-shirt with something on it, especially
0:05:13 if that something is recognizable.
0:05:15 So if I have Jimi Hendrix or Pink Floyd shirt or Johnny Cash or Metallica shirt, there’s
0:05:20 going to be certain people that look at me with recognition and respect and almost like
0:05:25 they want to start a conversation with me.
0:05:27 When I have a t-shirt, like a black t-shirt with nothing on it, that kind of look doesn’t
0:05:34 happen.
0:05:35 If I went out more, I would take a notebook and actually make this a little bit more
0:05:39 rigorous.
0:05:40 But anyway, there’s definitely a noticeable social effect that happens when you have
0:05:45 a t-shirt with a cool thing on it.
0:05:48 So I’m really happy with all the creators that are using Shopify to sell cool t-shirts.
0:05:52 I wish there was a better discovery process though.
0:05:56 I’m always in search of buying cool t-shirts.
0:05:59 Like I just, on Instagram, I think there was an advertisement for a set of t-shirts for
0:06:07 classic movies.
0:06:09 And that was really badass, but I scrolled past it and I regret it.
0:06:13 See that’s like a piece of advertisement that actually works.
0:06:17 But I wish there’s a way to not take me from the scrolling experience or maybe a way to
0:06:22 bookmark it really naturally.
0:06:24 There’s already a natural sort of skepticism about advertisement, but here it worked.
0:06:29 So like when advertisement is done well, it works.
0:06:32 I just wish I saved it.
0:06:34 But anyway, hopefully they use Shopify to sell shirts.
0:06:37 If they don’t, they should.
0:06:39 And if you’re thinking of selling shirts, use Shopify also.
0:06:43 Sign up for a $1 per month trial period at Shopify.com/Lux.
0:06:47 That’s all lowercase.
0:06:48 Go to Shopify.com/Lux to take your business to the next level today.
0:06:54 This episode is brought to you by BetterHelp, spelled H-E-L-B Help.
0:06:58 They figure out what you need to match you with a licensed therapist in under 48 hours.
0:07:03 Individuals, couples, the whole thing.
0:07:07 What are my favorite couples therapies in film?
0:07:09 I feel like Breaking Bad had good ones.
0:07:14 That’s a series, but I’m trying to think of it in a movie.
0:07:17 That’s a cool setting because I’ve been thinking about interviewing directors and actors more
0:07:22 and more and the setting of a couple’s therapy is really interesting.
0:07:29 It’s a really interesting dynamic between a man and a woman and a therapist and them
0:07:32 trying to sort of make explicit the implicit drama that’s been boiling over in their relationship.
0:07:40 Obviously there’s the therapist one-on-one relationship is really interesting on film.
0:07:45 Good Will Hunting with Robin Williams, man.
0:07:49 What a great, great performance.
0:07:52 I miss that guy so much.
0:07:54 What a truly special human being.
0:07:56 Anyway, back in the real world, therapy, even when there’s no camera, is really important.
0:08:03 Shining the light on the young in shadow, 350 million plus messages, 34,000 licensed
0:08:11 therapists, 4.4 million people who have gotten help through BetterHelp.
0:08:17 Check them out at BetterHelp.com/Lex and save on your first month at BetterHelp.com/Lex.
0:08:24 This episode is also brought to you by AG1, an all-in-one daily drink to support better
0:08:29 health and peak performance.
0:08:30 I drink it every day, multiple times a day, sometimes.
0:08:34 Usually after a run, like I’m going to go for a run in a little bit, it’s already that
0:08:38 Texas heat.
0:08:39 It’s warming up.
0:08:40 It’s warming up.
0:08:41 It’s creeping up on the 100 degree weather and I love it.
0:08:45 I don’t care.
0:08:46 The hotter it is, the tougher the run, the more of a mental test it is and what I do
0:08:52 is I speed up, take that feeling of discomfort and allow myself to sit in it and visualize
0:09:00 that feeling of discomfort fading.
0:09:03 From a third person perspective, it’s just a feeling and a feeling can be controlled.
0:09:08 A feeling can be ignored.
0:09:11 A feeling can be morphed from the negative to the positive.
0:09:15 For me, it’s not just a meditative practice of letting go of all feelings and focusing
0:09:19 on the breath.
0:09:20 For me, it is also being able to control that discomfort and letting go of that discomfort.
0:09:28 The feeling and the notion of discomfort, even when on the surface, there should be
0:09:32 a lot of physical discomfort because physical discomfort is first and foremost a construction
0:09:38 of the mind.
0:09:39 It’s not real.
0:09:40 It’s not real.
0:09:41 Because you believe it’s not real.
0:09:44 It’s not real.
0:09:45 And that’s what I do.
0:09:47 But when I get back home, extremely exhausted and uncomfortable, having overcome that challenge,
0:09:54 I put an AG1 in the freezer for like 30 minutes.
0:09:59 It has this great consistency and then after a shower, I just take the drink and celebrate
0:10:05 having overcome something difficult.
0:10:08 They’ll give you one month supply of fish oil.
0:10:10 Then you sign up at drinkag1.com/lex.
0:10:15 This is the Lex Freedom Podcast.
0:10:17 To support it, please check out our sponsors in the description.
0:10:20 And now, dear friends, here’s Sarah Walker.
0:10:40 We open the book, Life as No One Knows It, The Physics of Life’s Emergence with a distinction
0:10:46 between the materialists and the vitalists.
0:10:50 So what’s the difference?
0:10:51 Can you maybe define the two?
0:10:52 I think the question there is about whether life can be described in terms of matter and
0:11:03 physical things or whether there is some other feature that’s not physical that actually
0:11:11 animates living things.
0:11:13 So for a long time, people maybe have called that a soul.
0:11:17 It’s been really hard to pin down what that is.
0:11:19 So I think the vitalist idea is really that it’s kind of a dualistic interpretation that
0:11:24 there’s sort of the material properties, but there’s something else that animates life
0:11:29 that is there when you’re alive and it’s not there when you’re dead.
0:11:33 And materialists kind of don’t think that there’s anything really special about the matter
0:11:36 of life and the material substrates that life is made out of.
0:11:40 So they disagree on some really fundamental points.
0:11:43 Is there a gray area between the two?
0:11:47 Maybe all there is is matter, but there’s so much we don’t know that it might as well
0:11:51 be magic.
0:11:54 Out of that magic that the vitalists see, meaning there’s just so much mystery that
0:12:01 it’s really unfair to say that it’s boring and understood and as simple as “physics.”
0:12:08 Yeah, I think the entire universe is just a giant mystery.
0:12:13 I guess that’s what motivates me as a scientist.
0:12:15 And so oftentimes when I look at open problems like the nature of life or consciousness or
0:12:21 what is intelligence or are their souls or whatever question that we have that we feel
0:12:27 like we aren’t even on the tip of answering yet, I think we have a lot more work to do
0:12:33 to really understand the answers to these questions.
0:12:36 So it’s not magic, it’s just the unknown.
0:12:39 And I think a lot of the history of humans coming to understand the world around us has
0:12:43 been taking ideas that we once thought were magic or supernatural and really understanding
0:12:49 them in a much deeper way that we learn what those things are.
0:12:56 And they still have an air of mystery even when we understand them.
0:13:00 There’s no sort of bottom to our understanding.
0:13:04 So do you think the vitalists have a point that they’re more eager and able to notice
0:13:09 the magic of life?
0:13:12 I think that no tradition, vitalists included, is ever fully wrong about the nature of the
0:13:18 things that they’re describing.
0:13:20 So a lot of times when I look at different ways that people have described things across
0:13:25 human history, across different cultures, there’s always a seed of truth in them.
0:13:29 And I think it’s really important to try to look for those because if there are narratives
0:13:32 that humans have been telling ourselves for thousands of years, for thousands of generations,
0:13:38 there must be some truth to them.
0:13:40 We’ve been learning about reality for a really long time and we recognize the patterns that
0:13:46 reality presents us.
0:13:48 We don’t always understand what those patterns are and so I think it’s really important to
0:13:51 pay attention to that.
0:13:52 So I don’t think the vitalists were actually wrong and a lot of what I talk about in the
0:13:57 book but also I think about a lot just professionally is the nature of our definitions of what’s
0:14:03 material and how science has come to invent the concept of matter and that some of those
0:14:09 things actually really are inventions that happened in a particular time in a particular
0:14:14 technology that could learn about certain patterns and help us understand them and that
0:14:20 there are some patterns we still don’t understand and if we knew how to measure those things
0:14:27 or we knew how to describe them in a more rigorous way, we would realize that the material
0:14:33 world matter has more properties than we thought that it did and one of those might be associated
0:14:38 with the thing that we call life.
0:14:40 Life could be a material property and still have a lot of the features that the vitalists
0:14:44 thought were mysterious.
0:14:46 So we may still expand our understanding what is incorporated in the category of matter
0:14:52 that will eventually incorporate such magical things that the vitalists have noticed in
0:14:58 life.
0:14:59 Yeah, so I think about I always like to use examples from physics so I’ll probably do
0:15:03 that to like, like it’s just my go-to place but you know in the history of gravitational
0:15:11 physics for example in the history of motion, you know like when Aristotle came up with
0:15:15 his theories of motion, he did it by the material properties he thought things had.
0:15:19 So there was a concept of things falling to earth because they were solid like and things
0:15:24 raising to the heavens because they were air like and things moving around the planet because
0:15:28 they were celestial like but then we came to realize that thousands of years later and
0:15:33 after the invention of many technologies that allowed us to actually measure time in a mechanistic
0:15:39 way and track planetary motion and we could you know roll balls down incline planes and
0:15:46 track that progress.
0:15:47 We realized that if we just talked about mass and acceleration we could unify all motion
0:15:52 in the universe in a really simple description.
0:15:56 So we didn’t really have to worry about the fact that my cup is heavy and the air is light
0:16:00 like the same laws describe them if we have the right material properties to talk about
0:16:05 what those laws are actually interacting with and so I think the issue with life is we don’t
0:16:10 know how to think about information in a material way and so we haven’t been able to build a
0:16:16 unified description of what life is or the kind of things that evolution builds because
0:16:23 we haven’t really invented the right material concept yet.
0:16:27 So when talking about motion the laws of physics appear to be the same everywhere out in the
0:16:35 universe.
0:16:36 Do you think the same is true for other kinds of matter that we might eventually include
0:16:41 life in?
0:16:43 I think life obeys universal principles.
0:16:47 I think there is some deep underlying exploratory framework that will tell us about the nature
0:16:53 of life in the universe and will allow us to identify life that we can’t yet recognize
0:16:59 because it’s too different.
0:17:01 You’re right about the paradox of defining life.
0:17:04 Why does it seem to be so easy and so complicated at the same time?
0:17:09 All the sort of classic definitions people want to use just don’t work.
0:17:13 They don’t work in all cases.
0:17:15 So Carl Sagan had this wonderful essay on definitions of life where I think he talks
0:17:20 about aliens coming from another planet.
0:17:22 If they saw Earth they might think that cars were the dominant life form because there’s
0:17:26 so many of them on our planet and like humans are inside them.
0:17:30 You might want to exclude machines but any definition like classic biology textbook definitions
0:17:36 would also include them and so he wanted to draw a boundary between these kind of things
0:17:41 by trying to exclude them but they were naturally included by the definitions people want to
0:17:47 give.
0:17:48 What he ended up pointing out is that all of the definitions of life that we have, whether
0:17:52 it’s life is a self-reproducing system or life eats to survive or life requires compartments,
0:18:00 whatever it is, there’s always a counter example that challenges that definition.
0:18:04 This is why viruses are so hard or why fire is so hard and so we’ve had a really hard
0:18:09 time trying to pin down from a definitional perspective exactly what life is.
0:18:15 Yeah, you actually bring up the zombie ant fungus.
0:18:19 I enjoyed looking at this thing as an example of one of the challenges.
0:18:23 He mentioned viruses but this is a parasite.
0:18:26 Look at that.
0:18:27 Did you see this in the jungle?
0:18:29 Infects, ants.
0:18:30 Actually, one of the interesting things about the jungle, everything is ephemeral.
0:18:36 Everything eats everything really quickly.
0:18:38 So if an organism dies, that organism disappears, isn’t it?
0:18:43 Yeah.
0:18:44 It’s a machine that doesn’t have, I wanted to say it doesn’t have a memory or a history
0:18:50 which is interesting given your work on history in defining a living being.
0:18:57 The jungle forgets very quickly.
0:18:58 It wants to erase the fact that you existed very quickly.
0:19:01 Yeah, but it can’t erase it.
0:19:03 It’s just restructuring it.
0:19:04 I think the other thing that is really vivid to me about this example that you’re giving
0:19:08 is how much death is necessary for life.
0:19:12 So I worry a bit about notions of immortality and whether immortality is a good thing or
0:19:20 not.
0:19:21 So I have sort of a broad conception that life is the only thing the universe generates
0:19:26 that actually has even the potential to be immortal.
0:19:29 But that says this sort of process that you’re describing where life is about memory and
0:19:33 historical contingency and construction of new possibilities.
0:19:37 But when you look at any instance of life, especially one as dynamic as what you’re describing,
0:19:42 it’s a constant birth and death process.
0:19:44 But that birth and death process is like the way that the universe can explore what possibilities
0:19:51 can exist and not everything, not every possible human or every possible ant or every possible
0:19:58 zombie ant or every possible tree will ever live.
0:20:03 So it’s an incredibly dynamic and creative place because of all that death.
0:20:08 So does this thing, this is a parasite that needs the ant.
0:20:12 So is this a living thing or is this not a living thing?
0:20:15 So this is, it just pierces the ant.
0:20:18 I mean, and I’ve seen a lot of this, by the way, organisms working together in the jungle,
0:20:25 like ants protecting a delicious piece of fruit.
0:20:28 So they need the fruit, but like if you touch that fruit, they’re going to, like the forces
0:20:34 emerge.
0:20:35 They’re fighting you.
0:20:36 They’re defending that fruit to the death.
0:20:39 Just nature seems to find mutual benefits, right?
0:20:42 Yeah, it does.
0:20:44 I think the thing that’s perplexing for me about these kind of examples is effectively
0:20:49 the ants dead, but it’s staying alive now because it’s piloted by this fungus.
0:20:54 And so that gets back to this thing that we were talking about a few minutes ago about
0:20:58 how the boundary life is really hard to define.
0:21:00 So anytime that you want to draw a boundary around something and you say, this feature
0:21:06 is the thing that makes this alive or this thing is alive on its own, there’s not ever
0:21:12 really a clear boundary.
0:21:13 And these kind of examples are really good at showing that because it’s like the thing
0:21:17 that you would have thought is the living organism is now dead, except that it has another
0:21:22 living organism that’s piloting it.
0:21:23 So the two of them together are alive in some sense, but they’re now in this kind of weird
0:21:29 symbiotic relationship that’s taking the saint to its death.
0:21:32 So what do you do with that in terms of when you try to define life?
0:21:36 I think we have to get rid of the notion of an individual as being relevant.
0:21:40 And this is really difficult because a lot of the ways that we think about life, like
0:21:45 the fundamental unit of life is the cell.
0:21:49 Those are alive, but we don’t think about how gray that distinction is.
0:21:56 So for example, you might consider, you know, self-reproduction to be the most defining
0:22:03 feature of life.
0:22:04 A lot of people do actually like, you know, one of these standard different definitions
0:22:07 that a lot of people may feel like to use in astrobiology is life as a self-sustaining
0:22:10 chemical system capable of Darwinian evolution, which I was once quoted as agreeing with
0:22:15 and I was really offended because I hate that definition.
0:22:18 I think it’s terrible.
0:22:19 And I think it’s terrible that people use it.
0:22:22 I think like every word in that definition is actually wrong as a descriptor of life.
0:22:26 Life is a self-sustaining chemical system capable of Darwinian evolution.
0:22:29 Why is that?
0:22:30 That seems like a pretty good definition.
0:22:31 Yeah, I know.
0:22:32 If you want to make me angry, you can pretend I said that and believed it.
0:22:36 So self-sustaining chemical system, Darwinian evolution, what is self-sustaining?
0:22:44 It’s so frustrating.
0:22:45 I mean, which aspect is frustrating to you, but it’s also those very interesting words.
0:22:48 Yeah, they’re all interesting words.
0:22:50 And, you know, together, they sound really smart and they sound like they box in what
0:22:54 life is, but you can use any of the words individually and you can come up with counter-examples
0:23:00 that don’t fulfill that property.
0:23:02 The self-sustaining one is really interesting thinking about humans, right?
0:23:06 Like, we’re not self-sustaining, we’re dependent on societies.
0:23:09 And so, you know, I find it paradoxical that, you know, it might be that societies because
0:23:15 they’re self-sustaining units are now more alive than individuals are.
0:23:19 And that could be the case, but I still think we have some property associated with life.
0:23:23 I mean, that’s the thing that we’re trying to describe.
0:23:26 So that one’s quite hard and in general, you know, no organism is really self-sustaining.
0:23:31 They always require an environment.
0:23:33 So being self-sustaining is coupled in some sense to the world around you.
0:23:37 We don’t live in a vacuum.
0:23:41 So that part’s already challenging.
0:23:44 And then you can go to a chemical system.
0:23:47 I don’t think that’s good either.
0:23:48 I think there’s a confusion because life emerges in chemistry.
0:23:52 That life is chemical.
0:23:54 I don’t think life is chemical.
0:23:55 I think life emerges in chemistry because chemistry is the first thing the universe
0:24:00 builds where it cannot exhaust all the possibilities because the combinatorial space of chemistry
0:24:06 is too large.
0:24:07 Well, but is it possible to have a life that is not a chemical system?
0:24:10 Yes.
0:24:11 There’s a guy I know named Lee Cronin has been on a podcast a couple of times who just
0:24:14 got really pissed off.
0:24:15 I know what a coincidence.
0:24:17 He probably just got really pissed off hearing that.
0:24:20 I hope people somehow don’t know he’s a chemist.
0:24:22 Yeah, but he would agree with that statement.
0:24:25 Would he?
0:24:26 I don’t think he would.
0:24:27 I don’t think he would.
0:24:28 He would broaden the definition of chemistry until it would include everything.
0:24:31 Oh, sure.
0:24:32 Okay.
0:24:33 Or maybe.
0:24:34 I don’t know.
0:24:35 But I guess the first thing it creates is chemistry.
0:24:38 We’re very precisely, it’s not the first thing it creates.
0:24:41 Obviously, it has to make atoms first, but it’s the first thing.
0:24:44 If you think about the universe originated, atoms were made in Big Bang, nuclear synthesis,
0:24:50 and then later in stars, and then planets formed, and planets become engines of chemistry.
0:24:56 They start exploring what kind of chemistry is possible, and the combinatorial space of
0:25:04 chemistry is so large that even on every planet in the entire universe, you will never express
0:25:10 every possible molecule.
0:25:13 I like this example actually that Lee gave me, which is to think about taxol.
0:25:17 It has a molecular weight of about 853.
0:25:20 It’s got a lot of atoms, but it’s not astronomically large.
0:25:23 If you tried to make one molecule with that molecular formula in every three-dimensional
0:25:31 shape you could make with that molecular formula, it would fill.
0:25:34 It would fill 1.5 universes in volume with one unique molecule.
0:25:41 That’s just one molecule.
0:25:43 Chemical space is huge, and I think it’s really important to recognize that because if you
0:25:49 want to ask a question of why does life emerge in chemistry, well, life emerges in chemistry
0:25:53 because life is the physics of how the universe selects what gets to exist.
0:25:59 Those things get created along historically-continent pathways and memory and all the other stuff
0:26:03 that we can talk about, but the universe has to actually make historically-continent choices
0:26:08 in chemistry because it can exhaust all possible molecules.
0:26:11 What kind of things can you create that’s outside the combinatorial space of chemistry?
0:26:16 That’s what I’m trying to understand.
0:26:19 If it’s not chemical, so I think some of the things that have evolved on our biosphere,
0:26:24 I would call as much alive as chemistry, as a cell, but they seem much more abstract.
0:26:31 For example, I think language is alive, or at least life, I think memes are.
0:26:39 You’re saying language is life.
0:26:41 Language is alive.
0:26:42 Oh boy, I’m going to have to explore that one.
0:26:45 Life may be not alive, but I actually don’t know where I stand exactly on that.
0:26:51 I’ve been thinking about that a little bit more lately, but mathematics too.
0:26:56 It’s interesting because people think that math has this platonic reality that exists
0:27:00 outside of our universe, and I think it’s a feature of our biosphere, and it’s telling
0:27:05 us something about the structure of ourselves.
0:27:09 I find that really interesting because when you internalize all of these things that we
0:27:13 noticed about the world and you start asking, “Well, what do these look like if I was something
0:27:18 outside of myself observing these systems that we’re all embedded in?
0:27:23 What would that structure look like?”
0:27:25 I think we look really different than the way that we talk about what we look like to
0:27:29 each other.
0:27:30 What do you think a living organism in math is?
0:27:33 Is it one axiomatic system, or is it individual theorems, or is it–
0:27:37 I think it’s the fact that it’s open-ended in some sense.
0:27:44 It’s another open-ended combinatorial space, and the recursive properties of it allow creativity
0:27:51 to happen, which is what you see with the revolution in the last century with Gerdl’s
0:27:57 theorem and Turing, and there’s clear places where mathematics notices holes in the universe.
0:28:05 It seems like you’re sneaking up on a different kind of definition of life.
0:28:09 Open-ended, large combinatorial space, room for creativity.
0:28:15 Definitely not chemical.
0:28:16 I mean, chemistry is one substrate.
0:28:19 It’s restricted to chemical.
0:28:21 What about the third thing, which I think would be the hardest because you probably like it
0:28:25 the most is evolution or selection.
0:28:28 Well, specifically, it’s Darwinian evolution.
0:28:31 I think Darwinian evolution is a problem, but the reason that that definition is a problem
0:28:35 is not because evolution is in the definition, but because the implication is that most people
0:28:43 would want to make is that an individual is alive, and the evolutionary process, at least
0:28:48 the Darwinian evolutionary process, and most evolutionary processes, they don’t happen
0:28:53 at the level of individuals.
0:28:54 They happen at the level of populations.
0:28:56 Again, you would be saying something like what we saw with the self-sustaining definition,
0:29:00 which is that populations are alive, but individuals aren’t because populations evolve
0:29:05 and individuals don’t.
0:29:06 Obviously, maybe you’re alive because your gut microbiome is evolving, but Lex as an
0:29:12 entity right now is not evolving by canonical theories of evolution.
0:29:17 In assembly theory, which is attempting to explain life, evolution is a much broader
0:29:23 thing.
0:29:24 So, an individual organism can evolve under assembly theory?
0:29:27 Yes.
0:29:28 You’re constructing yourself all the time.
0:29:30 Assembly theory is about construction and how the universe selects for things to exist.
0:29:34 What if you reformulate everything like a population is a living organism?
0:29:38 That’s fine too, but this again gets back to it.
0:29:43 I think we can nitpick at definitions.
0:29:45 I don’t think it’s incredibly helpful to do it, but the reason for me-
0:29:49 It’s fun.
0:29:50 Yeah, it is fun.
0:29:51 It is really fun.
0:29:52 I do think it’s useful in the sense that when you see the ways that they all break down,
0:29:58 you either have to keep forcing in your conception of life you want to have or you have to say,
0:30:04 “All these definitions are breaking down for a reason.
0:30:06 Maybe I should adopt a more expansive definition that encompasses all the things that I think
0:30:12 are life.”
0:30:13 For me, I think life is the process of how information structures matter over time and
0:30:20 space.
0:30:21 An example of life is what emerges on a planet and yields an open-ended cascade of generation
0:30:28 of structure and increasing complexity.
0:30:31 This is the thing that life is.
0:30:33 Any individual is just a particular instance of these lineages that are structured across
0:30:40 time.
0:30:42 We focus so much on these individuals that are these short temporal moments in this larger
0:30:47 causal structure that actually is the life on our planet.
0:30:52 I think that’s why these definitions break down because they’re not general enough, they’re
0:30:57 not universal enough, they’re not deep enough, they’re not abstract enough to actually capture
0:31:00 that regularity.
0:31:01 Because we’re focused on those little affirmable things that you call a human life.
0:31:07 It’s like Aristotle focusing on heavy things falling because they’re Earth-like and things
0:31:13 floating because they’re air-like, it’s the wrong thing to focus on.
0:31:19 What exactly are we missing by focusing on such a short span of time?
0:31:23 I think we’re missing most of what we are.
0:31:26 One of the issues, I’ve been thinking about this really viscerally lately, it’s weird
0:31:31 when you do theoretical physics because I think it literally changes the structure of
0:31:34 your brain and you see the world differently, especially when you’re trying to build new
0:31:38 abstractions.
0:31:39 Do you think it’s possible if you’re a theoretical physicist that it’s easy to fall off the
0:31:43 cliff and go descend into madness?
0:31:46 I think you’re always on the edge of it, but I think what is amazing about being a scientist
0:31:53 and trying to do things rigorously is it keeps your sanity.
0:31:57 I think if I wasn’t a theoretical physicist, I would be probably not sane, but what it
0:32:03 forces you to do is hold the thought, you have to hold yourself to the fire of these
0:32:07 abstractions in my mind, have to really correspond to reality and I have to really test that
0:32:11 all the time.
0:32:13 I love building new abstractions and I love going to those incredibly creative spaces that
0:32:20 people don’t see as part of the way that we understand the world now, but ultimately
0:32:26 I have to make sure that whatever I’m pulling from that space is something that’s really
0:32:30 usable and really relates to the world outside of me.
0:32:33 That’s what science is.
0:32:35 We were talking about what we’re missing when we look at a small stretch of time and a small
0:32:41 stretch of space.
0:32:43 The issue is we evolve perception to see reality a certain way.
0:32:50 For us, space is really important and time feels fleeting.
0:32:55 I had a really wonderful mentor, Paul Davies, most of my career and Paul’s amazing because
0:33:00 he gives these little seed thought experiments all the time.
0:33:04 He used to ask me all the time when I was a postdoc, this was kind of a random tangent,
0:33:07 but it was like how much of the universe could be converted into technology if you were thinking
0:33:12 about long-term futures and stuff like that.
0:33:15 It’s a weird thought experiment, but there’s a lot of deep things there.
0:33:18 I do think a lot about the fact that we’re really limited in our interactions with reality
0:33:24 by the particular architectures that we evolved and so we’re not seeing everything and in
0:33:29 fact, our technology tells us this all the time because it allows us to see the world
0:33:33 in new ways by basically allowing us to perceive the world in ways that we couldn’t otherwise.
0:33:39 What I’m getting at with this is I think that living objects are actually huge.
0:33:47 There are some of the biggest structures in the universe, but they are not big in space,
0:33:50 they are big in time and we actually can’t resolve that feature.
0:33:54 We don’t interact with it on a regular basis, so we see them as these fleeting things that
0:33:58 have this really short temporal clock time without seeing how large they are.
0:34:03 When I’m saying time here, the way that people could picture it is in terms of causal structure.
0:34:08 If you think about the history of the universe to get to you and you imagine that that entire
0:34:14 history is you, that is the picture I have in my mind when I look at every living thing.
0:34:22 You have a tweet for everything.
0:34:26 You tweeted.
0:34:27 Doesn’t everyone?
0:34:28 You have a lot of poetic, profound tweets.
0:34:31 Thank you.
0:34:32 We have a lot of puzzles that take a long time to figure out.
0:34:36 You know what the trick is.
0:34:38 The reason they’re hard to write is because it’s compressing a very deep idea into a short
0:34:43 amount of space and I really like doing that intellectual exercise because I find it productive
0:34:46 for me.
0:34:47 Yeah.
0:34:48 It’s a very interesting kind of compression algorithm, though.
0:34:51 Yeah.
0:34:52 I like language.
0:34:53 I think it’s really fun to play with.
0:34:54 Yeah.
0:34:55 I wonder if AI can decompress it.
0:34:57 I would like to try this, but I think I use language in certain ways.
0:35:02 That are non-canonical and I do it very purposefully and it would be interesting to me how AI would
0:35:08 interpret it.
0:35:09 Yeah.
0:35:10 Your tweets would be a good touring test for our superintelligence.
0:35:13 Anyway, you tweeted that things only look emergent because we can’t see time.
0:35:21 So if we could see time, what would the world look like?
0:35:25 You’re saying you’ll be able to see everything that an object has been every step of the way
0:35:32 that led to this current moment and all the interactions that required to make that evolution
0:35:41 happen.
0:35:42 You would see this gigantic tail.
0:35:44 The universe is far larger in time than it is in space.
0:35:50 Yeah.
0:35:52 This planet is one of the biggest things in the universe.
0:35:55 Also the more complexity, the bigger the objects are.
0:35:58 Yeah.
0:35:59 I think the modern technosphere is the largest object in time in the universe that we know
0:36:05 about.
0:36:06 When you say technosphere, what do you mean?
0:36:08 I mean the global integration of life and technology on this planet.
0:36:14 All the technological things we’ve created?
0:36:16 Mm-hmm.
0:36:17 I don’t think of them as separate.
0:36:18 They’re very integrated with the structure that generated them.
0:36:22 You can almost imagine it.
0:36:23 Time is constantly bifurcating and it’s generating new structures.
0:36:28 These new structures are locally constructing the future.
0:36:34 Things like you and I are very close together in time because we didn’t diverge very early
0:36:40 in the history universe.
0:36:41 It’s very recent.
0:36:42 I think this is one of the reasons that we can understand each other so well and we can
0:36:46 communicate effectively.
0:36:48 I might have some sense of what it feels like to be you, but other organisms bifurcated from
0:36:55 us in time earlier.
0:36:57 This is just the concept of phylogeny, right?
0:37:00 If you take that deeper and you really think about that as the structure of the physics
0:37:05 that generates life and you take that very seriously, all of that causation is still bundled
0:37:13 up in the objects we observe today.
0:37:18 You and I are close in this temporal structure, but we’re also, we’re so close because we’re
0:37:25 really big and we only are very different and the most recent moments in the time that’s
0:37:31 embedded in us.
0:37:35 It’s hard to use words to visualize what’s in minds.
0:37:40 I have such a hard time with this sometimes.
0:37:42 Actually, I was thinking in the way over here, you have pictures in your brain and then they’re
0:37:48 hard to put into words, but I realized I always say I have a visual, but it’s not actually
0:37:53 I have a visual, I have a feeling because oftentimes I cannot actually draw a picture
0:37:56 in my mind for the things that I say, but sometimes they go through a picture before
0:38:02 they get to words, but I like experimenting with words because I think they help paint
0:38:06 pictures.
0:38:07 Yeah.
0:38:08 It’s again, some kind of compressed feeling that you can query to get a sense of the bigger
0:38:14 visualization that you have in mind.
0:38:16 Yeah.
0:38:17 It’s just a really nice compression.
0:38:20 I think the idea of this object that in it contains all the information about the history
0:38:27 of an entity that you see now, just trying to visualize that is pretty cool.
0:38:31 Yeah.
0:38:32 I mean, obviously the mind breaks down quickly as you step seconds and minutes back in time.
0:38:39 For sure.
0:38:41 I guess it’s just a gigantic object.
0:38:46 Yeah.
0:38:47 What are you supposed to be thinking about?
0:38:48 Yeah.
0:38:49 I think this is one of the reasons that we have such an ability to abstract as humans
0:38:54 because we are so gigantic that like the space that we can go back into is really large.
0:38:59 So like the more abstract you’re going, like the deeper you’re going in that space.
0:39:03 But in that sense, aren’t we fundamentally all connected?
0:39:06 Yes.
0:39:07 And this is why the definition of life cannot be the individual.
0:39:10 It has to be these lineages because they’re all connected or interwoven and they’re exchanging
0:39:14 parts all the time.
0:39:15 Yeah.
0:39:16 So maybe there are certain aspects of those lineages that can be lifelike.
0:39:20 They can be characteristics.
0:39:21 They can be measured like with the assembly theory that have more or less life.
0:39:25 But they’re all just fingertips of a much bigger object.
0:39:30 Yeah.
0:39:31 I think life is very high dimensional.
0:39:33 And in fact, I think you can be alive in some dimensions and not in others.
0:39:37 Like if you could project all the causation that’s in you in some features of you, very
0:39:44 little causation is required and like very little history.
0:39:47 And in some features a lot is.
0:39:50 So it’s quite difficult to take this really high dimensional, very deep structure and
0:39:57 project it into things that we really can understand and say like this is the one thing
0:40:04 that we’re seeing because it’s not one thing.
0:40:07 It’s funny we’re talking about this now and I’m slowly starting to realize one of the
0:40:11 things I saw when I took ayahuasca afterwards actually.
0:40:16 So the actual ceremony is four or five hours, but afterwards you’re still riding whatever
0:40:21 the thing that you’re riding and I got a chance to afterwards hang out with some friends and
0:40:28 just shoot the shit in the, you know, in the forest and I get to see their faces.
0:40:35 And what was happening with their faces and their hair is I would get this interesting
0:40:40 effect.
0:40:41 First of all, everything was beautiful and I just had so much love for everybody.
0:40:45 But I could see their past selves like behind them.
0:40:51 It was this effect where I guess it’s a blurring effect of where like if I move like this,
0:40:59 the faces that were just there are still there and it would just float like this, these behind
0:41:07 them, which will create this incredible effect.
0:41:09 But it’s also another way to think about that is I’m visualizing a little bit of that
0:41:15 object of the thing they wore just a few seconds ago.
0:41:18 It’s a cool little effect and now it’s like giving it a bit more profundity to the effect
0:41:25 that was just beautiful aesthetically, but it’s also beautiful from a physics perspective
0:41:32 because that is a past self, I get a little glimpse at the past self that they wore.
0:41:38 But then you take that to its natural conclusion, not just a few seconds ago, but just to the
0:41:45 beginning of the universe and you can probably get to that, get down that lineage.
0:41:51 It’s crazy that there’s billions of years inside all of us, all of us.
0:41:55 And then we connect obviously, not too long ago.
0:42:00 You mentioned just the techno sphere and you also wrote that the most alive thing on this
0:42:04 planet is our techno sphere.
0:42:07 Why is the technology we create a kind of life form?
0:42:10 Why are you seeing it as life?
0:42:13 Because it’s creative, but with us obviously, not independently of us and also because of
0:42:18 this lineage view of life.
0:42:20 And I think about life often as a planetary scale phenomena because that’s sort of the
0:42:24 natural boundary for all of this causation that’s bundled in every object in our biosphere.
0:42:30 And so for me, it’s just sort of the current boundary of how far life on our planet has
0:42:39 pushed into the things that our universe can generate.
0:42:43 And so it’s the furthest thing, it’s the biggest thing.
0:42:47 And I think a lot about the nature of life across different scales.
0:42:51 And so we have cells inside of us that are alive and we feel like we’re alive, but we
0:42:58 don’t often think about the societies that we’re embedded in as alive or global scale
0:43:04 organization of us and our technology on the planet as alive.
0:43:09 But I think if you have this deeper view into the nature of life, which I think is necessary
0:43:16 also to solve the original life, then you have to include those things.
0:43:20 All of them.
0:43:21 So you have to simultaneously think about life at every single scale.
0:43:26 Planetary and the bacteria level.
0:43:28 Yeah.
0:43:29 This is the hard thing about solving the problem of life, I think, is how many things you have
0:43:34 to integrate into building a sort of a unified picture of this thing that we want to call
0:43:40 life.
0:43:41 And a lot of our theories of physics are built on building deep regularities that explain
0:43:48 a really broad class of phenomena.
0:43:50 And I think we haven’t really traditionally thought about life that way.
0:43:53 But I think to get at some of these hardest questions, like looking for life on other
0:43:57 planets or the origin of life, you really have to think about it that way.
0:44:00 And so most of my professional work is just trying to understand every single thing on
0:44:06 this planet that might be an example of life, which is pretty much everything, and then trying
0:44:10 to figure out what’s the deeper structure underlying that.
0:44:13 Yeah.
0:44:14 Schrodinger wrote that living matter, while not alluding to laws of physics as established
0:44:18 up to date, is likely to involve other laws of physics, here they’re too unknown.
0:44:25 So to him…
0:44:27 I love that quote.
0:44:29 There was a sense that at the bottom of this are new laws of physics that could explain
0:44:36 this thing that we call life.
0:44:37 Yeah.
0:44:38 Schrodinger really tried to do what physicists try to do, which is explain things.
0:44:46 And his attempt was to try to explain life in terms of non-equilibrium physics, because
0:44:54 he thought that was the best description that we could generate at the time.
0:44:59 And so he did come up with something really insightful, which was to predict the structure
0:45:04 of DNA as an aperiodic crystal.
0:45:07 And that was for a very precise reason that was the only kind of physical structure that
0:45:11 could encode enough information to actually specify a cell.
0:45:14 We knew some things about genes, but not about DNA and its actual structure when he proposed
0:45:20 that.
0:45:21 But in the book, he tried to explain life as kind of going against entropy.
0:45:26 And so some people talked about it as like Schrodinger’s paradox, how can life persist
0:45:29 when the second law of thermodynamics is there.
0:45:33 But in open systems, that’s not so problematic.
0:45:35 And really the question is, why can life generate so much order?
0:45:40 And we don’t have a physics to describe that.
0:45:43 And it’s interesting, you know, generations of physicists have thought about this problem.
0:45:48 Oftentimes, it’s like when people are retiring, they’re like, “Oh, now I can work on life,”
0:45:53 or they’re like more senior in their career, and they’ve worked on other more traditional
0:45:56 problems.
0:45:57 And there’s still a lot of impetus in the physics community to think that non-equilibrium physics
0:46:01 will explain life, but I think that’s not the right approach.
0:46:05 I don’t think ultimately the solution to what life is is there, and I don’t really think
0:46:10 entropy has much to do with it unless it’s entirely reformulated.
0:46:14 Well, because you have to explain how interesting order, how complexity emerges from the soup.
0:46:20 Yes.
0:46:21 From randomness.
0:46:22 From randomness.
0:46:23 Physics currently can’t do that.
0:46:25 No.
0:46:26 Physics hardly even acknowledges that the universe is random at its base.
0:46:31 We like to think we live in a deterministic universe and everything’s deterministic, but
0:46:34 I think that’s probably, you know, an artifact of the way that we’ve written down laws of
0:46:40 physics since Newton invented modern physics and his conception of motion and gravity,
0:46:45 which, you know, he formulated laws that had initial conditions and fixed dynamical laws.
0:46:55 And that’s been sort of become the standard canon of how people think the universe works
0:46:59 and how we need to describe any physical system is with an initial condition and a law of
0:47:03 motion.
0:47:04 And I think that’s not actually the way the universe really works.
0:47:07 I think it’s a good approximation for the kind of systems that physicists have studied
0:47:11 so far.
0:47:12 And I think it will radically fail in the long term at describing reality at its more
0:47:20 basal levels, but I’m not saying there’s a base.
0:47:22 I don’t think that reality has a ground and I don’t think there’s a theory of everything,
0:47:27 but I think there are better theories and I think there are more explanatory theories
0:47:30 and I think we can get to something that explains much more than the current laws of
0:47:34 physics do.
0:47:35 When you say theory of everything, you mean like everything, everything.
0:47:39 Yeah, you know, like in physics right now, it’s really popular to talk about theories
0:47:42 of everything.
0:47:43 So string theory is supposed to be a theory of everything because it unifies quantum mechanics
0:47:46 and gravity.
0:47:48 And, you know, people have their different pet theories of everything and the challenge
0:47:53 with the theory of everything.
0:47:54 I really love this quote from David Krakauer, which is a theory of everything is a theory
0:47:59 of everything except those things that theorize.
0:48:00 Oh, you mean removing the observer from the thing?
0:48:04 Yeah.
0:48:05 So it’s also weird because if a theory of everything explained to everything, it should
0:48:08 also explain the theory.
0:48:09 So the theory has to be recursive and none of our theories of physics are recursive.
0:48:14 So it’s just a, it’s a, it’s a weird concept.
0:48:17 Yeah, but it’s very difficult to integrate the observer into a theory.
0:48:20 I don’t think so.
0:48:22 I think you can build a theory acknowledging that you’re an observer inside the universe.
0:48:26 But it doesn’t become recursive in that way.
0:48:28 And that’s, you’re saying it’s possible to make a theory that’s okay with that?
0:48:33 I think so.
0:48:34 I mean, I don’t think there’s always going to be the paradox of another metal level
0:48:39 you could build on the, the metal level, right?
0:48:41 So like if you assume this is your universe and you’re the observer outside of it, you
0:48:45 have some meta description of that universe, but then you need a meta description of you
0:48:49 describing that universe, right?
0:48:50 So, you know, this is one of the biggest challenges that we face being observers inside our universe.
0:48:57 And also, you know, why the paradoxes and the foundations of mathematics and any place
0:49:01 that we try to have observers in the system or a system describing itself show up.
0:49:08 But I think it is possible to build a physics that builds in those things intrinsically
0:49:13 without having them be paradoxical or have holes in the descriptions.
0:49:19 And so one, one place I think about this quite a lot, which I think can give you sort of
0:49:23 a more concrete example is, is the nature of like what we call fundamental.
0:49:28 So we typically define fundamental right now in terms of the smallest indivisible units
0:49:36 of matter.
0:49:37 So again, you have to have a definition of what you think material is and matter is.
0:49:40 But right now that, you know, what’s fundamental or elementary particles?
0:49:44 And we think they’re fundamental because we can’t break them apart further.
0:49:47 And obviously we have theories like string theory that if they’re right, would replace
0:49:51 the current description of what’s the most fundamental thing in our universe by replacing
0:49:55 with something smaller.
0:49:58 But we can’t get to those theories because we’re technologically limited.
0:50:01 And so if you, if you look at this from a historical perspective and you think about
0:50:08 explanations changing as physical systems like us learn more about the reality in which
0:50:16 they live, we once considered atoms to be the most fundamental thing.
0:50:21 And you know, it literally comes from the word indivisible.
0:50:24 And then we realized atoms had substructure because we built better technology, which
0:50:27 allowed us to quote unquote see the world better and resolve smaller features of it.
0:50:32 And then we built even better technology, which allowed us to see even smaller structure
0:50:36 and get down to the standard model particles.
0:50:39 And we think that there’s might be structure below that, but we can’t get there yet with
0:50:43 our technology.
0:50:44 So what’s fundamental, the way we talk about it in current physics is not actually fundamental.
0:50:53 It’s the boundaries of what we can observe in our universe, what we can see with our
0:50:57 technology.
0:50:58 And so if you want to build a theory that’s about us and about what’s inside the universe
0:51:06 that we can observe, not what’s at the boundary of it, you need to talk about objects that
0:51:12 are in the universe that you can actually break apart to smaller things.
0:51:16 So I think the things that are fundamental are actually the constructed objects.
0:51:19 They’re the ones that really exist and you really understand their properties because
0:51:22 you know how the universe constructed them because you can actually take them apart.
0:51:25 You can understand the intrinsic laws that built them.
0:51:28 But the things at the boundary are just at the boundary.
0:51:30 They’re evolving with us and we’ll learn more about that structure as we go along.
0:51:34 But really, if we want to talk about what’s fundamental inside our universe, we have to
0:51:37 talk about all these things that are traditionally considered emergent, but really just structures
0:51:42 in time that have causal histories that constructed them and are really actually what our universe
0:51:50 is about.
0:51:51 So we should focus on the construction methodology as the fundamental thing.
0:51:55 But do you think there’s a bottom to the smallest possible thing that makes up the universe?
0:52:01 I don’t see one.
0:52:03 And it’ll take way too long.
0:52:04 It’ll take longer to find that than it will to understand the mechanism that created life.
0:52:09 I think so, yeah.
0:52:10 I think for me, the frontier in modern physics, where the new physics lies, is not in high
0:52:16 energy particle physics.
0:52:17 It’s not in quantum gravity.
0:52:20 It’s not in any of these sort of traditionally sold.
0:52:22 This is going to be the newest, deepest insight we have into the nature of reality.
0:52:25 It is going to be in studying the problems of life and intelligence and the things that
0:52:30 are sort of also our current existential crises as a civilization or a culture that’s going
0:52:36 through an existential trauma of inventing technologies that we don’t understand right
0:52:42 now.
0:52:43 The existential trauma and the terror we feel that that technology might somehow destroy
0:52:48 us, us meaning living intelligently with organisms, yet we don’t understand what that
0:52:53 even means.
0:52:54 Well, humans have always been afraid of our technologies, though, right?
0:52:56 So it’s kind of a fascinating thing that every time we invent something we don’t understand,
0:53:01 it takes us a little while to catch up with it.
0:53:02 I think also in part, humans kind of love being afraid.
0:53:06 Yeah, we love being traumatized.
0:53:08 It’s weird.
0:53:09 We want to learn more, and then when we learn more, it traumatizes us.
0:53:14 You know, I never thought about it this before, but I think this is one of the reasons I love
0:53:17 what I do is because it traumatizes me all the time.
0:53:20 It sounds really bad, but what I mean is I love the shock of realizing that coming to
0:53:26 understand something in a way that you never understood it before.
0:53:30 I think it seems to me when I see a lot of the ways other people react to new ideas that
0:53:35 they don’t feel that way intrinsically, but for me, that’s why I do what I do.
0:53:39 I love that feeling.
0:53:42 But you’re also working on a topic where it’s fundamentally ego-destroying, because you’re
0:53:48 talking about life, and it’s humbling to think that the individual human is not special.
0:53:56 Yeah.
0:53:57 And you’re very viscerally exploring that.
0:54:00 Yeah, I’m trying to embody that because I think you have to live the physics to understand
0:54:07 it, but there’s a great quote about Einstein.
0:54:10 I don’t know if this is true or not, that he once said that he could feel like a beam
0:54:13 in his belly, but I think you got to think about it though, right?
0:54:20 If you’re a really deep thinker and you’re really thinking about reality that deeply,
0:54:23 and you are part of the reality that you’re trying to describe, you feel it.
0:54:27 You really feel it.
0:54:28 That’s what I was saying about, you’re always walking along the cliff.
0:54:32 If you fall off, you’re falling into madness.
0:54:35 Yes, it’s a constant, constant descent in the madness.
0:54:38 The fascinating thing about physicist in madness is that you don’t know if you’ve fallen off
0:54:43 the cliff.
0:54:44 Yeah, you know you don’t know.
0:54:45 That’s the cool thing about it.
0:54:46 I rely on other people to tell me.
0:54:47 Actually, this is very funny, because I have these conversations with my students often.
0:54:51 They’re worried about going crazy, and I have to reassure them that one of the reasons
0:54:57 they’ll stay sane is by trying to work on concrete problems.
0:55:02 Going crazy or waking up, I don’t know which one it is.
0:55:06 So what do you think is the origin of life on earth, and how can we talk about it in
0:55:12 a productive way?
0:55:13 The origin of life is this boundary that the universe can only cross if a structure that
0:55:22 emerges can reinforce its own existence, which is self-reproduction, auto-catalysis, things
0:55:27 people traditionally talk about, but it has to be able to maintain its own existence against
0:55:33 this sort of randomness that happens in chemistry, and this randomness that happens in the quantum
0:55:39 world.
0:55:40 In some sense, it’s the emergence of a deterministic structure that says, “I’m going to exist,
0:55:45 and I’m going to keep going,” but pinning that down is really hard.
0:55:50 We have ways of thinking about it in assembly theory that I think are pretty rigorous, and
0:55:53 one of the things I’m really excited about is trying to actually quantify in an assembly
0:55:59 theoretic way when the origin of life happens.
0:56:02 The basic process I have in mind is a system that has no causal contingency, no constraints
0:56:11 of objects basically constraining the existence of other objects or allowing the existence
0:56:17 of other objects.
0:56:19 That sounds very abstract, but you can just think of a chemical reaction can’t happen
0:56:23 if there’s not a catalyst, for example, or a baby can’t be born if there wasn’t a parent.
0:56:29 There’s a lot of causal contingency that’s necessary for certain things to happen.
0:56:35 You think about this unconstrained random system.
0:56:39 There’s nothing that reinforces the existence of other things, so the resources just get
0:56:45 washed out in all of these different structures, and none of them exist again, or they’re not
0:56:51 very complicated if they’re in high abundance.
0:56:55 Some random events allow some things to start reinforcing the existence of a small subset
0:57:01 of objects.
0:57:02 If they can do that, like just molecules basically recognizing each other and being able to catalyze
0:57:09 certain reactions, there’s this transition point that happens where unless you get a
0:57:19 self-reinforcing structure, something that can maintain its own existence, it actually
0:57:24 can’t cross this boundary to make any objects in high abundance without having this past
0:57:32 history that it’s carrying with us and maintaining the existence of that past history.
0:57:37 That boundary point where objects can’t exist unless they have this selection in history
0:57:41 in them is what we call the original life.
0:57:43 Pretty much everything beyond that boundary is holding on for dear life to all of the
0:57:49 causation and causal structure that’s basically put it there, and it’s carving its way through
0:57:54 this possibility space into generating more and more structure.
0:58:00 That’s when you get the open-ended cascade of evolution, but that boundary point is really
0:58:04 hard to cross.
0:58:05 Then what happens when you cross that boundary point and the way objects come into existence
0:58:08 is also really fascinating dynamics because as things become more complex, the assembly
0:58:14 index increases.
0:58:15 I can explain all these things.
0:58:16 Sorry, you can tell me what you want to explain or what people will want to hear.
0:58:22 Sorry, I have a very vivid visual on my brain, and it’s really hard to articulate it.
0:58:28 Got to convert it to language.
0:58:29 I know.
0:58:30 It’s so hard.
0:58:32 It’s going from a feeling to a visual to language is so stifling sometimes.
0:58:37 I have to convert it from language to a visual to a feeling.
0:58:42 Yeah.
0:58:43 I think it’s working.
0:58:44 I hope so.
0:58:46 I really like the self-reinforcing of the objects.
0:58:50 Just so I understand, one way to create a lot of the same kind of object is make them
0:58:55 self-reinforcing.
0:58:56 Yes.
0:58:57 So, self-reproduction has its property, right?
0:59:01 If the system can make itself, then it can persist in time, right?
0:59:06 Because all objects decay, they all have a finite lifetime, so if you’re able to make
0:59:10 a copy of yourself before you die, before the second law eats you or whatever people
0:59:16 think happens, then that structure can persist in time.
0:59:20 So that’s a way to sort of emerge out of a random soup, out of the randomness of soup.
0:59:26 Right.
0:59:27 But things that can copy themselves are very rare, and so what ends up happening is that
0:59:33 you get structures that enable the existence of other things, and then somehow, only for
0:59:43 some sets of objects, you get closed structures that are self-reinforcing and allow that entire
0:59:48 structure to persist.
0:59:49 Right.
0:59:50 So, the one object A reinforces the existence of object B, but object A can die.
0:59:58 Yeah.
0:59:59 So, you have to close that loop.
1:00:00 Right.
1:00:01 It’s just all very unlikely statistically, but that’s sufficiently, so you’re saying
1:00:10 there’s a chance.
1:00:11 There is a chance.
1:00:12 There’s no probability, but once you solve that, once you close the loop, you can create
1:00:16 a lot of those objects.
1:00:18 And that’s what we’re trying to figure out is what are the causal constraints that close
1:00:20 the loop.
1:00:21 So there is this idea that’s been in the literature for a really long time that was originally
1:00:24 proposed by Stuart Kaufman as really critical to the origin life called autocatalytic set.
1:00:28 So autocatalytic set is exactly this property.
1:00:30 We have A makes B, B makes C, C makes A, and you get a closed system.
1:00:34 But the problem with the theory of autocatalytic sets is incredibly brittle as a theory, and
1:00:39 it requires a lot of ad hoc assumptions like you have to assume function.
1:00:45 You have to say this thing makes B. It’s not an emergent property, the association between
1:00:50 A and B.
1:00:51 And so the way I think about it is much more general if you think about these histories
1:00:58 that make objects, it’s kind of like the structure of the histories collapses in such a way that
1:01:06 these things are all in the same sort of causal structure, and that causal structure actually
1:01:11 loops back on itself to be able to generate some of the things that make the higher level
1:01:15 structures.
1:01:16 Lee has a beautiful example of this actually in molybdenum.
1:01:19 It’s like the first non-organic, autocatalytic set.
1:01:24 It’s a self-reproducing molybdenum ring, but it’s like molybdenum.
1:01:31 And basically, if you look at the molybdenum, it makes a huge molybdenum ring.
1:01:35 I don’t remember exactly how big it is.
1:01:36 It might be like 150 molybdenum atoms or something.
1:01:39 But if you think about the configuration space of that object, it’s exponentially large.
1:01:43 How many possible molecules?
1:01:44 So why does the entire system collapse on just making that one structure?
1:01:49 If you start from molybdenum atoms that are maybe just like a couple of them stuck together.
1:01:54 And so what they see in this system is there’s a few intermediate stages.
1:01:58 So there’s like some random events where the chemistry comes together and makes these structures.
1:02:02 And then once you get to this very large one, it becomes a template for the smaller ones.
1:02:05 And then the whole system just reinforces its own production.
1:02:08 How did Lee find this molybdenum close loop?
1:02:12 If I knew how Lee’s brainwork, I think I would understand more about the universe, but I…
1:02:20 This is not an algorithmic discovery, it’s a…
1:02:22 No, but I think it goes to the deepest roots of when he started thinking about origins of life.
1:02:28 So I don’t know all his history, but what he’s told me is he started out in crystallography.
1:02:33 And there are some things that people would just take for granted about chemical structures
1:02:42 that he was deeply perplexed about, just like why are these really intricate,
1:02:47 really complex structures forming so easily under these conditions.
1:02:50 And he was really interested in life, but he started in that field.
1:02:56 So he’s just carried with him these deep insights from these systems
1:02:59 that seem like they’re totally not alive and just these metallic chemistries
1:03:04 into actually thinking about the deep principles of life.
1:03:08 So I think he already knew a lot about that chemistry, and he also…
1:03:15 Assembly theory came from him thinking about how these systems work.
1:03:21 So he had some intuition about what was going on with this molybdenum ring.
1:03:26 The molybdenum might be able to be the thing that makes a ring.
1:03:30 They knew about them for a long time, but they didn’t know that the mechanism
1:03:33 of why that particular structure form was autocatalytic feedback.
1:03:37 And so that’s what they figured out in this paper.
1:03:40 And I actually think that paper is revealing some of the mechanism
1:03:43 of the origin of life transition.
1:03:44 Because really what you see, like the origin of life is basically like,
1:03:49 you should have a combinatorial explosion of the space of possible structures
1:03:54 that are too large to exhaust.
1:03:56 And yet you see it collapse on this really small space of possibilities
1:04:02 that’s mutually reinforcing itself to keep existing.
1:04:06 That is the origin of life.
1:04:08 There’s some set of structures that result in this autocatalytic feedback.
1:04:13 Yeah.
1:04:14 And what is that, tiny, tiny, tiny, tiny percent?
1:04:17 I think it’s a small space, but chemistry is very large.
1:04:22 So there might be a lot of them out there, but we don’t know.
1:04:27 And one of them is the thing that probably started life on earth.
1:04:29 That’s right.
1:04:30 Or many, many starts.
1:04:32 Yes.
1:04:32 They keep starting maybe.
1:04:33 Yeah.
1:04:34 I mean, there’s also all kinds of other weird properties that happen
1:04:37 around this kind of phase boundary.
1:04:41 So this other project that I have in my lab is focused on the origin of chirality,
1:04:46 which is thinking about– so chirality is this property of molecules
1:04:52 that they can come in mirror image forms.
1:04:54 So like chiral learning means hand.
1:04:56 So your left and right hand are what’s called non-superimposable
1:05:00 because if you try to lay one on the other,
1:05:02 you can’t actually lay them directly on top of each other.
1:05:04 And that’s the property of being a mirror image.
1:05:08 So there’s this sort of perplexing property of the chemistry life
1:05:10 that no one’s been able to really adequately explain
1:05:13 that all of the amino acids and proteins are left-handed
1:05:17 and all of the bases in RNA and DNA are right-handed.
1:05:22 And yet the chemistry of these building block units,
1:05:27 the amino acids and nucleobases is the same for left and right-handed.
1:05:30 So you have to have some kind of symmetry breaking
1:05:32 where you go from these chemistries that seem entirely equivalent
1:05:35 to only having one chemistry takeover as the dominant form.
1:05:40 And for a long time, I had been really–
1:05:43 I actually did my PhD on the origin of chirality.
1:05:46 I was working on it as a symmetry breaking problem in physics.
1:05:50 This is how I got started in the origin of life.
1:05:52 And then I left it for a long time
1:05:53 because I thought it was one of the most boring problems in the origin of life.
1:05:55 But I’ve come back to it because I think there’s something really deep going on here
1:05:58 related to this combinatorial explosion of the space of possibilities.
1:06:02 But just to get to that point, this feature of this handedness
1:06:07 has been the main focus.
1:06:08 But people take for granted the existence of chiral molecules at all,
1:06:14 that this property of having a handedness.
1:06:17 And they just assume that it’s just a generic feature of chemistry.
1:06:23 But if you actually look at molecules,
1:06:26 if you look at chemical space, which is the space of all possible molecules
1:06:29 that people can generate, and you look at small molecules,
1:06:33 things that have less than about 7 to 11 heavy atoms,
1:06:37 so things that are not hydrogen,
1:06:39 almost every single molecule in that space is a chiral,
1:06:42 like doesn’t have a chiral center.
1:06:43 So it would be like a spoon.
1:06:45 A spoon doesn’t have a– it’s the same as its mirror image.
1:06:48 It’s not like a hand that’s different than its mirror image.
1:06:51 But if you get to this threshold boundary, above that boundary,
1:06:56 almost every single molecule is chiral.
1:06:59 So you go from a universe where almost nothing has a mirror image form.
1:07:03 There’s no mirror image universe of possibilities to this one
1:07:06 where every single structure has pretty much a mirror image version.
1:07:10 And what we’ve been looking at in my lab is that it seems to be the case that
1:07:17 the original life transition happens around the time when you start accumulating.
1:07:22 You push your molecules to a large enough complexity
1:07:26 that chiral molecules become very likely to form.
1:07:29 And then there’s a cascade of molecular recognition
1:07:33 where chiral molecules can recognize each other.
1:07:36 And then you get this sort of autocatalytic feedback
1:07:38 and things self-reinforcing.
1:07:39 So is chirality in itself an interesting feature or just an accident of complexity?
1:07:45 No, it’s a super interesting feature.
1:07:46 I think chirality breaks symmetry and time, not space.
1:07:49 So we think of it as a spatial property, like a left and right hand.
1:07:53 But if I choose the left hand,
1:07:55 I’m basically choosing the future of that system for all time
1:07:58 because I basically made a choice between the ways
1:08:01 that that molecule can now react with every other object in its chemical universe.
1:08:05 Oh, I see.
1:08:06 And so you’re actually like,
1:08:07 when you have the splitting of making a molecule
1:08:10 that now has another form it could have had by the same exact atomic composition,
1:08:16 but now it’s just a mirror image isometry.
1:08:17 You’re basically splitting the universe of possibilities every time.
1:08:20 Yeah, in two.
1:08:23 In two, but molecules can have more than one chiral center,
1:08:25 and that’s not the only stereosometry that they can have.
1:08:27 So this is one of the reasons that taxol fills 1.5 universes of space.
1:08:32 It’s all of these spatial permutations that you do on these objects
1:08:35 that actually makes the space so huge.
1:08:36 So the point of this sort of chiral transition that I’m pointing out
1:08:41 is chirality is actually a signature of being in a complex chemical space.
1:08:46 And the fact that we think it’s a really generic feature of chemistry
1:08:49 and it’s really prevalent is because most of the chemistry we study on Earth
1:08:53 is a product already of life.
1:08:54 And it also has to do with this transition in assembly,
1:08:57 this transition in possibility spaces,
1:08:59 because I think there’s something really fundamental going on at this boundary
1:09:03 that you don’t really need to go that far into chemical space
1:09:08 if you can to actually see life in terms of this depth in time,
1:09:12 this depth in symmetries of objects in terms of chiral symmetries
1:09:17 or this assembly structure.
1:09:19 But getting past this boundary that’s not very deep in that space requires life.
1:09:26 It’s a really weird property.
1:09:31 And it’s really weird that so many abrupt things happen in chemistry at that same scale.
1:09:35 So would that be the greatest invention ever made on Earth
1:09:41 in its evolutionary history?
1:09:43 So I really like that formulation of it.
1:09:45 Nick Lane has a book called Life Ascending
1:09:48 where he lists the 10 great inventions of evolution,
1:09:51 the origin of life being first and DNA,
1:09:54 the hereditary material that encodes the genetic constructions for all living organisms,
1:09:59 then photosynthesis, the process that allows organisms to convert sunlight
1:10:04 into chemical energy producing oxygen as a byproduct,
1:10:07 the complex cell, eukaryotic cells,
1:10:10 which contain a nucleus and organelles that rose from simple bacterial cells,
1:10:14 sex, sexual reproduction, movement.
1:10:18 So just the ability to move under which you have the predation,
1:10:21 the predators and ability of living organisms.
1:10:24 I like that movement in there.
1:10:25 That’s cool.
1:10:26 Yeah, but a movement includes a lot of interesting stuff in there,
1:10:29 like predator prey dynamic, which not to romanticized a nature as metal.
1:10:35 That seems like the important one.
1:10:37 I don’t know.
1:10:37 It’s such a computationally powerful thing to have a predator and prey.
1:10:44 Well, it’s efficient for things to eat other things that are already alive
1:10:47 because they don’t have to go all the way back to the base chemistry.
1:10:51 Well, that, but maybe I just like deadlines,
1:10:54 but it creates an urgency.
1:10:55 You’re going to get eaten.
1:10:56 You got to live.
1:10:58 Yeah, like survival.
1:10:59 It’s not just the static and private you’re battling against.
1:11:03 You’re like the dangers against which you’re trying to survive are also evolving.
1:11:10 This is just a much faster way to explore this base of possibilities.
1:11:15 I actually think it’s a gift that we don’t have much time.
1:11:18 Yes, a sight, the ability to see.
1:11:21 So the increasing, complexifying of sensory organisms, consciousness and death,
1:11:28 the concept of programmed cell death.
1:11:31 These are all inventions along the line.
1:11:36 I like invention as a word for them.
1:11:38 I think that’s good.
1:11:38 Which are the more interesting inventions to you?
1:11:41 Well, the origin of life, because you kind of are not glorifying the origin of life itself.
1:11:47 There’s a process.
1:11:47 No, I think the origin of life is a continual process.
1:11:50 That’s why I’m interested in the first transition and solving that problem
1:11:53 because I think it’s the hardest, but I think it’s happening all the time.
1:11:57 When you look back at the history of earth, what do you impress happened?
1:12:01 I like sight as an invention because I think having sensory perception
1:12:10 and trying to comprehend the world to use anthropocentric terms is a really critical
1:12:17 feature of life.
1:12:18 It’s interesting the way that sight has complexified over time.
1:12:24 So if you think of the origin of life, nothing on the planet could see.
1:12:30 So for a long time, life had no sight.
1:12:33 And then photon receptors were invented.
1:12:37 And then when multicellularity evolved, those cells eventually grew into eyes.
1:12:44 And we had the multicellular eye.
1:12:47 And then it’s interesting when you get to societies, like human societies,
1:12:51 that we invent even better technologies of seeing telescopes and microscopes,
1:12:55 which allow us to see deeper into the universe or at smaller scales.
1:13:00 So I think that’s pretty profound the way that sight has transformed the ability of life
1:13:08 to literally see the reality in which it’s existing.
1:13:13 I think consciousness is also obviously deeply interesting.
1:13:21 I’ve gotten obsessed with octopus.
1:13:26 They’re just so weird and the fact that they evolved complex nervous systems
1:13:31 kind of independently seems very alien.
1:13:33 Yeah, there’s a lot of alien organisms.
1:13:36 That’s another thing I saw in the jungle.
1:13:38 Yeah.
1:13:40 Just things that are like, oh, OK, they make one of those, huh?
1:13:43 Do you have any examples?
1:13:46 There’s a frog that’s as thin as a sheet of paper.
1:13:51 And I was like, what?
1:13:53 And it gets birthed through pores.
1:13:55 Oh, I’ve seen videos of that.
1:13:57 It’s so gross when the babies come out.
1:13:59 Did you see that in person, like the babies coming out?
1:14:02 Oh, no, no.
1:14:03 I saw the without the–
1:14:05 Have you seen videos of that?
1:14:06 It’s so gross.
1:14:07 It’s one of the grossest things I’ve ever seen.
1:14:09 Well, gross is just the other side of beautiful.
1:14:13 It’s like, oh, wow, that’s possible.
1:14:17 I guess if I was one of those frogs,
1:14:19 I would think that was the most beautiful event I’d ever seen.
1:14:21 Although like human childbirth is not that beautiful either.
1:14:24 Yeah, it’s all out of perspective.
1:14:27 Well, we come into the world so violently.
1:14:29 It’s just like, it’s amazing.
1:14:31 Well, I mean, the world is a violent place.
1:14:33 Yeah.
1:14:34 So again, another–
1:14:35 It’s just another side of the coin.
1:14:38 You know what?
1:14:38 This actually makes me think of one that’s not up there,
1:14:40 which I do find really incredibly amazing is the process
1:14:48 of the germline cell in organisms.
1:14:54 Like basically every living thing on this planet,
1:14:56 at some point in its life, has to go through a single cell.
1:14:59 And this whole issue of development.
1:15:01 Like the developmental program is kind of crazy.
1:15:03 Like how do you build you out of a single cell?
1:15:05 How does a single cell know how to do that?
1:15:07 Pattern formation of a multicellular organism
1:15:11 obviously evolves with DNA.
1:15:13 But there’s a lot of stuff happening there
1:15:15 about when cells take on certain morphologies
1:15:18 and things that people don’t understand
1:15:19 like the actual shape formation mechanism.
1:15:21 And a lot of people study that.
1:15:23 And there’s a lot of advances being made now in that field.
1:15:27 I think it’s pretty shocking, though,
1:15:28 that like how little we know about that process.
1:15:30 And often it’s left off of people’s lists.
1:15:33 It’s just kind of interesting.
1:15:34 Mbrio genesis is fascinating.
1:15:36 Yeah, because you start from just one cell.
1:15:39 Yeah.
1:15:40 And the genes in all the cells are the same, right?
1:15:41 So like the differentiation has to be
1:15:44 something that’s like much more about like the actual like,
1:15:49 you know, expression of genes over time
1:15:54 and like how they get switched on and off
1:15:55 and also the physical environment
1:15:57 of like the cell interacting with other cells.
1:15:59 And there’s just a lot of stuff going on.
1:16:02 Yeah, the computation, the intelligence of that process.
1:16:05 Yes.
1:16:06 Might be like the most important thing to understand
1:16:09 and we just kind of don’t really think about it.
1:16:11 Right.
1:16:12 We think about the final product.
1:16:13 Yeah.
1:16:13 Maybe the key to understanding the organism
1:16:18 is understanding that process, not the final product.
1:16:21 Probably, yes.
1:16:22 I think most of the things about understanding anything
1:16:25 about what we are embedded in time.
1:16:27 Well, of course you would say that.
1:16:28 I know.
1:16:29 So predictable.
1:16:29 It’s turning into a deterministic universe.
1:16:35 Always has been.
1:16:36 Always was like the meme.
1:16:38 Yeah, always was, but it won’t be in the future.
1:16:40 Well, that’s before we talk about the future,
1:16:42 let’s talk about the past, the assembly theory.
1:16:44 Can you explain assembly theory to me?
1:16:48 I listened to Lee talk about it for many hours
1:16:50 and I understood nothing.
1:16:51 No, I’m just kidding.
1:16:52 I just wanted to take another,
1:16:54 you’ve been already talking about it,
1:16:56 but just from a big picture view
1:17:03 is the assembly theory way of thinking about our world.
1:17:10 About our universe.
1:17:12 Yeah.
1:17:12 I think the first thing is the observation that life seems
1:17:21 to be the only thing in the universe that builds complexity
1:17:24 and the way that we see it here.
1:17:25 And complexity is obviously like a loaded term.
1:17:28 So I’ll just use assembly instead
1:17:30 because I think assembly is more precise.
1:17:32 But the idea that all the things on your desk here
1:17:38 from your computer to the pen to us sitting here
1:17:43 don’t exist anywhere else in the universe
1:17:45 as far as we know, they only exist on this planet.
1:17:47 And it took a long evolutionary history to get to us.
1:17:50 Is a real feature that we should take seriously
1:17:54 as one that’s deeply embedded in the laws of physics
1:17:58 and the structure of the universe that we live in.
1:18:00 Standard physics would say that all of that complexity
1:18:03 traces back to the infinitesimal.
1:18:08 Deviations and like the initial state of the universe
1:18:10 that there was some order there.
1:18:11 I find that deeply unsatisfactory.
1:18:14 And what assembly theory says that’s very different
1:18:19 is that the universe is basically constructing itself.
1:18:25 And when you get to these common historical spaces
1:18:29 like chemistry where the space of possibilities
1:18:33 is too large to exhaust them all, you can only construct things
1:18:39 along historically contingent paths.
1:18:42 Like you basically have causal chains of events
1:18:44 that happen to allow other things to come into existence.
1:18:47 That this is the way that complex objects get formed
1:18:53 is basically on scaffolding on the past history
1:18:55 of objects making more complex objects,
1:18:57 making more complex objects.
1:18:59 That idea in itself is easy to state and simple,
1:19:02 but it has some really radical implications
1:19:04 as far as what you think is the nature of the physics
1:19:10 that would describe life.
1:19:11 And so what assembly theory does formally
1:19:14 is try to measure the boundary
1:19:16 in the space of all things that chemically could exist,
1:19:21 for example, like all possible molecules.
1:19:23 Where is the boundary above which we should say
1:19:25 these things are too complex to happen
1:19:27 outside of an evolutionary chain of events,
1:19:30 outside of selection.
1:19:31 And we formalize that with two observables.
1:19:35 One of them is the copy number of the object.
1:19:37 So how many of the object did you observe?
1:19:39 And the second one is what’s the minimal number
1:19:41 of recursive steps to make it.
1:19:43 So if you start from elementary building blocks
1:19:48 like bonds for molecules and you put them together
1:19:51 and then you take things you’ve made already
1:19:53 and build up to the object,
1:19:54 what’s the shortest number of steps you had to take?
1:19:57 And what Lee’s been able to show in the lab with his team
1:20:01 is that for organic chemistry, it’s about 15 steps.
1:20:07 And then you only see molecules that,
1:20:10 you know, the only molecules that we observe
1:20:14 that are past that threshold are ones that are in life.
1:20:18 And in fact, one of the things I’m trying to do
1:20:19 with this idea of like trying to actually quantify
1:20:21 the original life as a transition
1:20:23 in like a phase transition assembly theory
1:20:26 is actually be able to explain why that boundary is where it is.
1:20:31 Because I think that’s actually the boundary
1:20:32 that life must cross.
1:20:34 So the idea of going back to this thing
1:20:36 we were talking about before about these structures
1:20:38 that can reinforce their own existence
1:20:40 and move past that boundary,
1:20:41 15 seems to be that boundary in chemical space.
1:20:44 It’s not a universal number.
1:20:46 It will be different for different assembly spaces.
1:20:48 But that’s what we’ve experimentally validated so far.
1:20:52 And then–
1:20:53 So literally 15, like the assembly index is 15?
1:20:56 It’s 15 or so for the experimental data, yeah.
1:20:58 So that’s when you start getting the self-reinforcing.
1:21:01 That’s when you have to have that feature
1:21:04 in order to observe molecules in high abundance in that space.
1:21:09 So the copy number is the number of exact copies.
1:21:12 That’s what we mean by high abundance.
1:21:13 And assembly index or the complexity of the object is
1:21:17 how many steps it took to create it, recursive.
1:21:20 Recursive, yeah.
1:21:22 So you can think of objects in assembly theory
1:21:24 as basically recursive stacks
1:21:26 of the construction steps to build them.
1:21:28 So they’re like, it’s like you take this step
1:21:32 and then you make this object
1:21:33 and you make it this object and make this object
1:21:35 and then you get up to the final object.
1:21:36 But that object is all of that history
1:21:38 rolled up into the current structure.
1:21:40 What if you took the long way home?
1:21:41 You can’t take the long way.
1:21:43 Why not?
1:21:44 The long way doesn’t exist.
1:21:45 It’s a good song, though.
1:21:47 What do you mean the long way doesn’t exist?
1:21:50 If I do a random walk from A to B,
1:21:53 I’ll eventually, if I start an A,
1:21:55 I’ll eventually end up at B
1:21:57 and that random walk would be much shorter
1:21:59 than the longer than the shorter.
1:22:01 If you look at objects and you,
1:22:03 so we define something we call the assembly universe
1:22:06 and the assembly universe is ordered in time.
1:22:08 It’s actually ordered in the causation.
1:22:10 The number of steps to produce an object.
1:22:12 And so all objects in the universe
1:22:15 are in some sense existed a layer
1:22:18 that’s defined by their assembly index.
1:22:21 And the size of each layer is growing exponentially.
1:22:24 So what you’re talking about,
1:22:27 if you want to look at the long way of getting to an object,
1:22:30 as I’m increasing the assembly index of an object,
1:22:32 I’m moving deeper and deeper
1:22:33 into an exponentially growing space.
1:22:35 And it’s actually also the case
1:22:37 that the sort of typical path to get to that object
1:22:41 is also exponentially growing
1:22:42 with respect to the assembly index.
1:22:44 And so if you want to try to make
1:22:47 a more and more complex object
1:22:48 and you want to do it by a typical path,
1:22:52 that’s actually an exponentially receding horizon.
1:22:54 And so most objects that come into existence
1:22:57 have to be causally very similar
1:22:58 to the things that exist
1:22:59 because they’re close by in that space
1:23:00 and they can actually get to it
1:23:02 by an almost shortest path for that object.
1:23:03 – Yeah, the almost shortest path is the most likely.
1:23:06 And like, buy a lot.
1:23:09 – Buy a lot.
1:23:10 – Okay, so if you see a high copy number.
1:23:12 – Yeah, imagine yourself–
1:23:14 – A copy number greater than one.
1:23:15 – Yeah, I mean, basically we live,
1:23:17 the more complex we get,
1:23:18 we live in a space that is growing exponentially large
1:23:23 and the ways of getting to objects in the space
1:23:26 are also growing exponentially large.
1:23:28 And so we’re this kind of recursively stacked structure
1:23:34 of all of these objects
1:23:36 that are clinging onto each other for existence
1:23:39 and then they grab something else
1:23:41 and are able to bring that thing into existence
1:23:43 because it’s kind of similar to them.
1:23:45 – But there is a phase transition.
1:23:46 There is a–
1:23:47 – There is a transition.
1:23:48 – There is a place where you would say,
1:23:50 oh, that’s a life.
1:23:51 – I think it’s actually abrupt.
1:23:51 I’ve never been able to say that
1:23:53 in my entire career before.
1:23:54 I’ve always gone back and forth
1:23:55 about whether the original life was kind of gradual
1:23:57 or abrupt.
1:23:57 I think it’s very abrupt.
1:23:58 – Poetically, chemically, literally.
1:24:01 What snaps, okay, that’s very beautiful.
1:24:02 – It snaps.
1:24:03 We’ll be poetic today.
1:24:05 But no, I think there’s like a lot of random exploration
1:24:09 and then there’s like,
1:24:09 and then the possibility space just collapses
1:24:13 on the structure kind of really fast
1:24:15 that can reinforce its own existence
1:24:17 because it’s basically fighting against non-existence.
1:24:20 – Yeah, you tweeted,
1:24:22 “The most significant struggle for existence
1:24:25 in the evolutionary process
1:24:27 is not among the objects that do exist,
1:24:29 but between the ones that do
1:24:31 and those that never have the chance to.”
1:24:33 This is where selection does most of its causal work.
1:24:37 The objects that never get a chance to exist,
1:24:44 the struggle between the ones that never get a chance
1:24:45 to exist and the ones that, okay, what’s that line exactly?
1:24:49 – I don’t know, we can make songs out of all of these.
1:24:51 – What are the objects that never get a chance to exist?
1:24:54 What does that mean?
1:24:54 – So there was this website, I forgot what it was,
1:24:58 but it’s like a neural network
1:25:00 that just generates a human face.
1:25:02 And it’s like, this person does not exist.
1:25:04 I think that’s what it’s called, right?
1:25:05 So you can just click on that all day
1:25:06 and you can look at people all day that don’t exist.
1:25:08 All of those people exist
1:25:10 in that space of things that don’t exist.
1:25:13 – Yeah, but there’s the real struggle.
1:25:17 – Yeah, so the struggle of the quote,
1:25:19 the struggle for existence,
1:25:21 is that goes all the way back
1:25:22 to Darwin’s writing about natural selection, right?
1:25:25 So like the whole idea of survival of the fittest
1:25:27 is everything struggling to exist,
1:25:28 this predator-prey dynamic, and the fittest survive.
1:25:33 And so the struggle for existence
1:25:35 is really what selection is all about.
1:25:37 But you’re, and that’s true.
1:25:40 We do see things that do exist,
1:25:44 competing to continue to exist.
1:25:45 But each time that, like if you think about
1:25:50 this space of possibilities,
1:25:52 and each time the universe generates a new structure,
1:25:56 or like an object that exists
1:25:59 generates a new structure along this causal chain,
1:26:03 it’s generating something that exists
1:26:06 that never existed before.
1:26:07 And each time that we make that kind of decision,
1:26:10 we’re excluding a huge space of possibilities.
1:26:13 And so actually like as this process
1:26:15 of increasing assembly index,
1:26:16 it’s not just that like the space
1:26:19 that these objects exist in is exponentially growing,
1:26:22 but there are objects in that space
1:26:24 that are exponentially receding away from us.
1:26:28 So they’re becoming exponentially less
1:26:30 and less likely to ever exist.
1:26:31 And so existence excludes a huge number of things.
1:26:36 Just because of the accident of history,
1:26:39 how it ended up.
1:26:40 Yeah, it is in part an accident
1:26:42 because I think some of the structure
1:26:45 that gets generated is driven a bit by randomness.
1:26:50 I think a lot of it, so one of the conceptions
1:26:53 that we have in assembly theory
1:26:55 is the universe is random at its base.
1:26:57 You can see this in chemistry,
1:26:59 like unconstrained chemical reactions are pretty random.
1:27:01 And then, and also quantum mechanics,
1:27:06 there’s less places that give evidence for that.
1:27:08 And deterministic structures emerge
1:27:11 by things that can causally reinforce themselves
1:27:14 and maintain persistence over time.
1:27:16 And so we are some of the most deterministic things
1:27:20 in the universe.
1:27:21 And so like we can generate very regular structure
1:27:25 and we can generate new structure
1:27:27 along a particular lineage,
1:27:28 but the possibility space at the sort of tips,
1:27:31 like the things we can generate next is really huge.
1:27:34 So there’s some stochasticity
1:27:36 in what we actually instantiate
1:27:39 as like the next structures that get built in the biosphere.
1:27:43 It’s not completely deterministic
1:27:46 because the space of future possibilities
1:27:48 is always larger than the space of things that exist now.
1:27:51 So how many instantiations of life is out there, do you think?
1:27:55 So how often does this happen?
1:27:59 What we see happen here on Earth,
1:28:02 how often is this process repeated throughout our galaxy,
1:28:05 throughout the universe?
1:28:06 So I said before, like right now,
1:28:08 I think the original life is a continuous process on Earth.
1:28:10 Like I think this idea of like combinatorial spaces
1:28:14 that our biosphere generates,
1:28:15 not just chemistry, but other spaces,
1:28:17 often cross this threshold
1:28:19 where they then allow themselves to persist
1:28:23 with a particular regular structure over time.
1:28:24 So language is another one where, you know,
1:28:27 like the space of possible configurations
1:28:31 of the 26 letters of the English alphabet
1:28:33 is astronomically large,
1:28:34 but we use with very high regularity certain structures.
1:28:37 And then we associate meaning to them
1:28:40 because of the regularity of like how much we use them, right?
1:28:43 So meaning is an emergent property of the causation
1:28:46 and the objects and like how often they recur
1:28:48 and what the relationship of the recurrence is to other objects.
1:28:51 Meaning is the emergent property, okay, got it.
1:28:53 Well, this is why you can play with language so much actually.
1:28:56 So words don’t really carry meaning,
1:28:57 it’s just about how you lace them together.
1:28:59 Yeah, but from where does the…
1:29:02 But you don’t have a lot of room,
1:29:04 obviously as a speaker of a given language,
1:29:06 you don’t have a lot of room with a given word to wiggle,
1:29:09 but you do have a certain amount of room
1:29:13 to push the meanings of words.
1:29:15 And I do this all the time and you have to do it
1:29:18 with the kind of work that I do,
1:29:21 because if you want to discover an abstraction,
1:29:27 like some kind of concept that we don’t understand yet,
1:29:30 it means we don’t have the language.
1:29:31 And so the words that we have are inadequate to describe the things.
1:29:36 This is why we’re having a hard time talking about assembly theory
1:29:38 because it’s a newly emerging idea.
1:29:40 And so I’m constantly playing with words in different ways
1:29:45 to try to convey the meaning that is actually behind the words,
1:29:48 but it’s hard to do.
1:29:51 So you have to wiggle within the constraints?
1:29:53 Yes, lots of wiggle.
1:29:55 The great orators are just good at wiggling.
1:29:59 Do you wiggle?
1:30:01 I’m not a very good wiggler, no.
1:30:05 This is the problem, this is part of the problem.
1:30:07 No, I like playing with words a lot.
1:30:09 You know, it’s very funny because, you know, like,
1:30:12 I know you talked about this with Lee,
1:30:13 but like people are so offended by the writing of the paper
1:30:16 that came out last fall.
1:30:18 And it was interesting because the ways that we use words
1:30:21 were not the way that people were interacting with the words.
1:30:24 And I think that was part of the mismatch
1:30:27 where we were trying to use words in a new way
1:30:29 because we were trying to describe something that,
1:30:32 you know, hadn’t been described adequately before,
1:30:35 but we had to use the words that everyone else uses
1:30:37 for things that are related.
1:30:38 And so it was really interesting to watch that clash play out
1:30:42 in real time for me,
1:30:44 being someone that tries to be so precise with my word usage,
1:30:47 knowing that it’s always going to be vague.
1:30:49 Boy, can I relate.
1:30:51 What is truth?
1:30:55 Is truth the thing you meant when you wrote the words
1:30:58 or is truth the thing that people understood
1:31:00 when they read the words?
1:31:01 Oh, yeah.
1:31:02 I think that compression mechanism into language
1:31:05 is a really interesting one,
1:31:06 and that’s why Twitter is a nice exercise.
1:31:09 I love Twitter.
1:31:10 You get to write a thing,
1:31:11 and you think a certain thing when you write it,
1:31:16 and then you get to see all these other people interpret it
1:31:18 in all kinds of different ways.
1:31:20 I use it as an experimental platform for that reason.
1:31:22 I wish there was a higher diversity
1:31:25 of interpretation mechanisms applied to tweets,
1:31:29 meaning like all kinds of different people would come to it,
1:31:33 like some people that see the good in everything,
1:31:36 and some people that are ultra cynical,
1:31:38 a bunch of haters, and a bunch of lovers, and a bunch of–
1:31:40 Maybe they could do better jobs
1:31:42 with presenting material to people.
1:31:45 Like how things–
1:31:47 It’s usually based on interests,
1:31:48 but I think it would be really nice
1:31:49 if you got like 10% of your Twitter feed
1:31:52 was random stuff sampled from other places.
1:31:54 That’d be kind of fun.
1:31:55 True.
1:31:55 I also would love to filter just like been the response
1:32:02 to tweets by the people that hate on everything.
1:32:06 Yes.
1:32:06 The people that are–
1:32:07 Oh, that would be fantastic.
1:32:08 The people that are super positive on everything,
1:32:11 and they’ll just kind of, I guess, normalize their response,
1:32:16 because then it’d be cool to see if the people
1:32:18 that are usually positive about everything
1:32:19 are hating on you, or like totally don’t understand
1:32:22 or completely misunderstood.
1:32:24 Yeah, usually it takes a lot of clicking to find that out.
1:32:26 Yeah, so it’d be better if it was sorted, yeah.
1:32:29 The more clicking you do,
1:32:30 the more damaging it is to the soul.
1:32:34 Yeah, it’s like instead of like–
1:32:36 You could have the blue check,
1:32:36 but you should have like,
1:32:37 are you a pessimist, an optimist,
1:32:40 there’s a lot of colors.
1:32:41 Geotic neutral.
1:32:41 Yeah, a whole rainbow of checks.
1:32:45 And then you realize there’s more categories
1:32:48 than we can possibly express in colors.
1:32:50 Yeah, of course.
1:32:51 People are complex.
1:32:53 That’s our best feature.
1:32:57 I don’t know how we got to the wiggling required,
1:33:01 given the constraints of language,
1:33:04 because I think we started about me asking about alien life,
1:33:10 which is how many different times
1:33:14 did the face transition happen elsewhere?
1:33:18 Do you think there’s other alien civilizations out there?
1:33:21 This goes into like the,
1:33:23 are you on the boundary of insane or not?
1:33:25 But when you think about the structure
1:33:29 of the physics of what we are that deeply,
1:33:31 it really changes your conception of things.
1:33:33 And going to this idea of the universe,
1:33:39 you know, being kind of small in physical space
1:33:43 compared to how big it is in time and like how large we are,
1:33:46 it really makes me question about whether there’s any other
1:33:50 structure that’s like this giant crystal in time,
1:33:53 this giant causal structure,
1:33:55 like our biosphere slash technosphere is
1:33:58 anywhere else in the universe.
1:34:00 Why not?
1:34:02 I don’t know.
1:34:04 Just because this one is gigantic,
1:34:06 this doesn’t mean there’s other giant.
1:34:08 Right, but I think the universe is expanding,
1:34:12 right?
1:34:12 It’s expanding in space,
1:34:13 but in assembly theory, it’s also expanding in time.
1:34:15 And actually that’s driving the expansion in space.
1:34:19 And expansion in time is also driving the expansion
1:34:23 in the sort of combinatorial space of things on our planet.
1:34:27 So that’s driving the sort of, you know,
1:34:29 pace of technology and all the other things.
1:34:31 So time is driving all of these things,
1:34:32 which is a little bit crazy to think
1:34:34 that the universe is just getting bigger
1:34:36 because time is getting bigger.
1:34:38 But like the sort of visual that gets built in my brain
1:34:43 about that is like the structure that we’re building
1:34:46 on this planet is packing more and more time
1:34:49 in this very small volume of space, right?
1:34:52 Because our planet hasn’t changed its physical size
1:34:54 in four billion years,
1:34:55 but there’s like a ton of causation and recursion and time,
1:35:00 whatever word you want to use,
1:35:02 information packed into this.
1:35:04 And I think this is also, you know, embedded
1:35:08 in sort of the virtualization of our technologies
1:35:11 or the abstraction of language and all of these things.
1:35:14 These things that seem really abstract
1:35:16 are just really deep in time.
1:35:18 And so what that looks like is you have a planet
1:35:24 that becomes increasingly virtualized.
1:35:27 And so it’s getting bigger and bigger in time,
1:35:29 but not really expanding out in space.
1:35:30 And the rest of space is like kind of moving away from it.
1:35:33 It’s again, it’s a sort of exponentially receding horizon.
1:35:35 And I’m just not sure how far into this evolutionary process
1:35:39 something gets if it can ever see
1:35:41 that there’s another such structure out there.
1:35:44 What do you mean by virtualized in that context?
1:35:46 Virtual as sort of a play on virtual reality
1:35:49 and like simulation theories.
1:35:51 But virtual also in a sense of, you know,
1:35:54 we talk about virtual particles and particle physics,
1:35:59 which, you know, they are very critical to doing calculations
1:36:02 about predicting the properties of real particles,
1:36:04 but we don’t observe them directly.
1:36:05 So what I mean by virtual here is virtual reality for me,
1:36:12 things that appear virtual, appear abstract,
1:36:15 are just things that are very deep in time
1:36:17 in the structure of the things that we are.
1:36:21 So if you think about you as a four billion year old object,
1:36:24 the things that are part of you,
1:36:27 like your capacity to use language or think abstractly
1:36:29 or have mathematics are just very, you know,
1:36:32 like deep temporal structures.
1:36:34 That’s why they look like they’re informational
1:36:37 and abstract is because they’re like,
1:36:39 they’re existing in this temporal part of you,
1:36:41 but not necessarily spatial part.
1:36:43 Just because I have a four billion year old history,
1:36:45 why does that mean I can’t hang out with aliens?
1:36:48 There’s a couple ideas that are embedded here.
1:36:50 So one of them comes again from Paul.
1:36:52 He wrote this book years ago about, you know,
1:36:56 like the eerie silence and why we’re alone.
1:36:58 And he concluded the book
1:36:59 with this idea of quintaligence or something,
1:37:01 but like this idea that like really advanced intelligence
1:37:05 would basically just build itself into a quantum computer.
1:37:09 And it would want to operate in the vacuum of space
1:37:12 because that’s the best place to do quantum computation.
1:37:14 And it would just like run out all of its computations
1:37:16 indefinitely, but it would look completely dark
1:37:18 to the rest of the universe.
1:37:19 And I don’t think as it’s typical,
1:37:21 like I don’t think that’s actually like the right physics,
1:37:23 but I think something about that idea
1:37:25 as I do with all ideas is partially correct.
1:37:27 And Freeman Dyson also had this amazing paper
1:37:30 about how long life could persist in a universe
1:37:33 that was exponentially expanding.
1:37:35 And his conception was like,
1:37:36 if you imagine analog life form,
1:37:38 it could run slower and slower and slower and slower
1:37:43 and slower as a function of time.
1:37:45 And so it would be able to run indefinitely
1:37:48 even against an exponentially expanding universe
1:37:51 because it would just run exponentially slower.
1:37:53 And so I guess part of what I’m doing in my brain
1:37:56 is putting those two things together
1:37:58 along with this idea that we are building,
1:38:01 you know, like if you imagine with our technology,
1:38:05 we’re now building virtual realities, right?
1:38:07 Like things we actually call virtual reality,
1:38:10 which required four billions of years of history
1:38:13 and a whole bunch of data to basically embed them
1:38:15 in a computer architecture.
1:38:16 So now you can put like an Oculus headset on
1:38:19 and think that you’re in this world, right?
1:38:21 And what you really are embedded in
1:38:23 is in a very deep temporal structure.
1:38:25 And so it’s huge in time, but it’s very small in space.
1:38:29 And you can go lots of places in the virtual space, right?
1:38:32 But you’re still stuck in like your physical body
1:38:34 and like sitting in the chair.
1:38:36 And so, you know, part of it is,
1:38:38 it might be the case that sufficiently evolved
1:38:40 biospheres kind of virtualize themselves.
1:38:46 And they internalize their universe
1:38:48 in their sort of temporal causal structure
1:38:50 and they close themselves off from the rest of the universe.
1:38:53 I just don’t know if a deep temporal structure
1:38:55 necessarily means that you’re closed off.
1:38:57 No, I don’t either.
1:38:58 So that’s kind of my fear.
1:39:00 So I’m not sure I’m agreeing with what I say.
1:39:03 I’m just saying like this is one sort of conclusion.
1:39:05 And, you know, like in my most sort of like,
1:39:07 it’s interesting because I don’t do psychedelic drugs.
1:39:10 But when people describe to me like your thing
1:39:12 with the faces and stuff and like I have,
1:39:14 you know, had a lot of deep conversations
1:39:15 with friends that have done psychedelic drugs
1:39:17 for intellectual reasons and otherwise.
1:39:19 But I’m always like, oh, it sounds like
1:39:22 you’re just doing theoretical physics.
1:39:23 Like that’s what brains do on theoretical physics.
1:39:25 So I live in these like really abstract spaces most of the time.
1:39:30 But there’s also this issue of extinction, right?
1:39:34 Like, extinction events are basically pinching off
1:39:37 an entire like causal structure than one of these like,
1:39:39 I’m going to call them time crystals.
1:39:41 I don’t like know it,
1:39:42 but there’s like these very large objects in time,
1:39:44 pinching off that whole structure from the rest of it.
1:39:46 And so it’s like, if you imagine that sort of same thing
1:39:49 in the universe, I, you know,
1:39:51 I once thought that sufficiently advanced technologies
1:39:54 would look like black holes.
1:39:55 So it’d be just completely imperceptible.
1:39:57 Yeah. So, so there might be lots of aliens out there.
1:40:01 Maybe that’s the explanation for all the singularities.
1:40:03 They’re all pinched off causal structures
1:40:05 that virtualize the reality and kind of broke off from us.
1:40:07 Black holes in every way.
1:40:09 So like untouchable to us or unlikely to be
1:40:13 detectable by us with whatever sensory mechanisms we have.
1:40:18 Yeah. But the other way I think about it is,
1:40:20 is there is probably hopefully life out there.
1:40:24 So like I do work on life detection efforts in the solar system.
1:40:28 And I’m trying to help with the habitable world’s observatory
1:40:31 mission planning right now.
1:40:32 And working with like the biosignatures team for that,
1:40:35 like to think about exoplanet biosignatures.
1:40:37 So like I have some optimism that we might find things.
1:40:41 But there are the challenges
1:40:45 that we don’t know the likelihood for life,
1:40:47 like, which is what you were talking about.
1:40:48 So if I get to a more grounded discussion,
1:40:51 what I’m really interested in doing
1:40:53 is trying to solve the origin of life.
1:40:57 So we can understand how likely life is out there.
1:41:00 So I don’t think that the,
1:41:01 I think that the problem of discovering alien life
1:41:05 and solving the origin of life are deeply coupled
1:41:08 and in fact are one in the same problem.
1:41:10 And that the first contact with the alien life
1:41:13 will actually be in an original life experiment.
1:41:16 But that part I’m super interested in.
1:41:18 And then there’s this other feature that I think about a lot,
1:41:21 which is our own technological phase of development
1:41:25 as sort of like what is this phase
1:41:27 in the evolution of life on a planet.
1:41:31 If you think about a biosphere emerging on a planet
1:41:33 and evolving over billions of years
1:41:35 and evolving into a technosphere.
1:41:37 When a technosphere can move off planet
1:41:41 and basically reproduce itself on another planet,
1:41:45 now you have biospheres reproducing themselves.
1:41:49 Basically they have to go through technology to do that.
1:41:52 And so there are ways of thinking about sort of
1:41:57 the nature of intelligent life
1:41:58 and how it spreads in that capacity
1:42:00 that I’m also really excited about and thinking about.
1:42:03 And all of those things for me are connected.
1:42:07 Like we have to solve the origin of life
1:42:08 in order for us to get off planet
1:42:10 because we basically have to start life on another planet.
1:42:13 And we also have to solve the origin of life
1:42:15 in order to recognize other alien intelligence.
1:42:17 Like all of these things are like literally the same problem.
1:42:20 – Right, understanding the origin of life here on earth
1:42:22 is a way to understand ourselves into.
1:42:25 Understanding ourselves is a prerequisite
1:42:27 for me to detect other intelligent civilizations.
1:42:32 I for one, take it for what it’s worth,
1:42:36 I in ayahuasca, one of the things I did is zoom out,
1:42:41 like aggressively, like a spaceship.
1:42:44 And it would always go quickly through the galaxy
1:42:47 and from the galaxy to this representation of the universe.
1:42:53 And at least for me, from that perspective,
1:42:55 it seemed like it was full of alien life.
1:42:57 Not just alien life, but intelligent life.
1:43:01 – I like that.
1:43:03 – And conscious life.
1:43:04 So like I don’t know how to convert it into words.
1:43:08 It’s more like a feeling, like you were saying.
1:43:09 – Yeah.
1:43:10 – A feeling converted to a visual to convert to words.
1:43:13 So I had a visual with it, but really it was a feeling
1:43:18 that it was just full of this vibrant energy.
1:43:22 That I was feeling when I’m looking at the people in my life
1:43:26 and full of gratitude, but that same exact thing
1:43:30 is everywhere in the universe.
1:43:33 – Right.
1:43:33 – So.
1:43:34 – I totally agree with this, like that visual I really love.
1:43:37 And I think we live in a universe
1:43:40 that like generates life and purpose
1:43:43 and like it’s part of the structure of just the world.
1:43:47 And so maybe like this sort of lonely view I have is,
1:43:51 I never thought about it this way,
1:43:53 so you’re describing that.
1:43:54 I was like, I want to live in that universe
1:43:55 and I’m like a very optimistic person
1:43:56 and I love building visions of reality that are positive.
1:44:01 But I think for me right now in the intellectual process,
1:44:04 I have to tunnel through this particular way
1:44:06 of thinking about the loneliness
1:44:09 of being like separated in time from everything else,
1:44:13 which I think like we also all are
1:44:15 because time is what defines us as individuals.
1:44:17 – So part of you is drawn to the trauma of being alone.
1:44:20 – Yeah.
1:44:21 – Deep in the physics.
1:44:23 – Yeah, but also part of what I mean
1:44:25 is like you have to go through ideas
1:44:27 you don’t necessarily agree with
1:44:30 to work out what you’re trying to understand.
1:44:32 And I’m trying to be inside this structure
1:44:34 so I can really understand it.
1:44:36 And I don’t think I’ve been able to like,
1:44:37 I’m so deeply embedded in what we are intellectually right now
1:44:42 that I don’t have an ability to see these other ones
1:44:46 that you’re describing if they’re there.
1:44:48 – Well, one of the things you kind of described
1:44:49 that you already spoke to,
1:44:51 you call it the great perceptual filter.
1:44:53 – Yeah.
1:44:53 – So there’s the famous great filter,
1:44:56 which is basically the idea that there’s some really powerful
1:45:02 moment in every intelligence civilization
1:45:06 that where they destroy themselves.
1:45:08 – Yeah.
1:45:09 – That explains why we have not seen aliens.
1:45:12 And you’re saying that there’s something like that
1:45:15 in the temporal history of the creation of complex objects
1:45:18 that at a certain point they become an island,
1:45:22 an island too far to reach based on the perceptions.
1:45:25 – I hope not, but yeah, I worry about it, yeah.
1:45:28 – But that’s basically meaning
1:45:31 there’s something fundamental about the universe
1:45:33 where if the more complex you become,
1:45:35 the harder it will be to perceive other complex creatures.
1:45:38 – Yeah.
1:45:38 I mean, just think about us with microbial life, right?
1:45:40 Like we used to once be cells.
1:45:42 And for most of human history,
1:45:44 we didn’t even recognize how your life was there
1:45:46 until we built a new technology,
1:45:48 microscopes that allowed us to see them.
1:45:49 Right, so that’s kind of, it’s kind of weird, right?
1:45:53 Like, like things that we-
1:45:54 – And they’re close to us.
1:45:55 – They’re close, they’re everywhere.
1:45:57 – But also in the history of the development
1:45:59 of complex objects, they’re pretty close.
1:46:01 – Yeah, super close, super close.
1:46:03 Like, yeah, I mean, everything on this planet is like,
1:46:07 it’s like pretty much the same thing.
1:46:10 Like the space of possibilities is so huge.
1:46:13 It’s like we’re virtually identical.
1:46:15 – So how many flavors or kinds of life do you think are possible?
1:46:20 – I’m kind of like trying to imagine
1:46:22 all the little flickering lights in the universe,
1:46:24 like in the way that you were describing,
1:46:25 that was kind of cool.
1:46:25 – It was so, I mean, it was obvious to me.
1:46:27 It was exactly that.
1:46:29 It was like lights.
1:46:29 The way you maybe see a city, but a city from like up above,
1:46:36 you see a city with the flickering lights,
1:46:38 but there’s a coldness to the city.
1:46:39 There’s some, you know, that humans are capable of good and evil,
1:46:44 and you could see like, there’s a complex feeling to the city.
1:46:47 I had no such complex feeling about seeing the lights
1:46:51 of all the galaxies, whatever, the billions of galaxies.
1:46:55 – Yeah, this is kind of cool.
1:46:56 I’ll answer the question in a second,
1:46:57 but maybe like this idea of flickering lights and intelligence
1:47:00 is interesting to me because I, you know,
1:47:02 like we have such a human centric view of alien intelligences
1:47:07 that a lot of the work that I’ve been doing with my lab
1:47:09 is just trying to take inspiration from non-human life on Earth.
1:47:15 And so I have this really talented undergrad student
1:47:18 that’s basically building a model of alien communication
1:47:22 based on fireflies.
1:47:24 So one of my colleagues or a peleg is, she’s totally brilliant,
1:47:28 but she goes out with like GoPro cameras and like,
1:47:30 you know, films in high resolution,
1:47:32 all these firefly flickering.
1:47:33 And she has like this theory about how their signaling evolved
1:47:36 to like maximally differentiate the flickering pattern.
1:47:41 So like she has a theory basically that predicts,
1:47:44 you know, like this species should flash like this.
1:47:46 If this one’s flashing like this,
1:47:47 this other one’s going to do it at a slower rate
1:47:49 so that the, you know, like they can distinguish each other
1:47:52 living in the same environment.
1:47:54 And so this undergrad’s building this model
1:47:56 where you have like a pulsar background
1:47:58 of all these like giant flashing sources in the universe
1:48:00 and an alien intelligence, you know,
1:48:02 wants to signal it’s there.
1:48:03 So it’s flashing like a firefly.
1:48:05 And I just like, I like the idea of thinking
1:48:08 about non-human aliens.
1:48:10 So that was really fun.
1:48:11 The mechanism of the flashing, unfortunately,
1:48:13 is like the diversity of that is very high.
1:48:15 And we might not be able to see it.
1:48:16 That’s what.
1:48:17 Yeah. Well, I think there’s some ways
1:48:19 we might be able to differentiate that signal.
1:48:20 I’m still thinking about this part of it.
1:48:22 So one is like, like if you have pulsars
1:48:25 and they all have a certain spectrum
1:48:27 to their pulsing patterns,
1:48:28 and you have this one signal that’s in there
1:48:30 that’s basically tried to maximally differentiate itself
1:48:33 from all the other sources in the universe,
1:48:35 it might stick out in the distribution.
1:48:37 Like there might be ways of actually being able to tell
1:48:38 if it’s an anomalous pulsar, basically.
1:48:41 But I don’t know if that would really work or not.
1:48:44 So still thinking about it.
1:48:45 You tweeted,
1:48:47 “If one wants to understand how truly combinatorially
1:48:49 and compositionally complex our universe is,
1:48:51 they only need step into the world of fashion.
1:48:55 Yeah.
1:48:56 It’s bonkers.
1:48:57 Total bonkers.
1:48:57 How big the constructable space of human aesthetics is.
1:49:02 Can you explain?
1:49:03 Can we explore the space of human aesthetics?
1:49:06 Yeah, I don’t know.
1:49:08 I’ve been kind of obsessed with that.
1:49:10 I never know how to pronounce it.
1:49:12 It’s a chopper rally.
1:49:13 Like, you know, like they have ears and things.
1:49:16 Like it’s such like a weird grotesque aesthetic.
1:49:18 But like, it’s totally bizarre.
1:49:20 But what I meant, like I have a visceral experience
1:49:24 when I walk into my closet.
1:49:25 I have like a lot of…
1:49:27 How big is your closet?
1:49:29 It’s pretty big.
1:49:30 It’s like I do assembly theory every morning
1:49:33 when I walk in my closet because I have,
1:49:35 I really like a very large combinatorial diverse palette,
1:49:39 but I never know what I’m going to build in the morning.
1:49:41 Do you get rid of stuff?
1:49:42 Sometimes.
1:49:43 Or do you have trouble getting rid of stuff?
1:49:45 Like…
1:49:45 I have trouble getting rid of some stuff.
1:49:47 It depends on what it is.
1:49:48 If it’s vintage, it’s hard to get rid of
1:49:50 because it’s kind of hard to replace.
1:49:52 It depends on the piece.
1:49:54 Yeah.
1:49:55 So you have your closet.
1:49:57 Is that one of those temporal time crystals that…
1:50:00 Yeah.
1:50:00 They just, you get to visualize the entire history of the…
1:50:03 It’s a physical manifestation of my personality.
1:50:06 Right.
1:50:06 So why is that a good visualization of the combinatorial
1:50:12 and compositionally complex?
1:50:15 I think it’s an interesting feature of our species
1:50:18 that we allow, we get to express ourself through what we wear.
1:50:21 Right.
1:50:22 Like if you think about all those animals in the jungle you saw,
1:50:25 like they’re born looking the way they look,
1:50:27 and then they’re stuck with it for life.
1:50:29 That’s true.
1:50:29 I mean, it is one of the loudest, clearest, most consistent ways
1:50:33 we signal to each other is the clothing we wear.
1:50:36 Yeah.
1:50:36 And it’s highly dynamic.
1:50:39 I mean, you can be dynamic if you want to.
1:50:41 Very few people are, there’s a certain bravery,
1:50:44 but it’s actually more about confidence,
1:50:46 willing to play with style and play with aesthetics.
1:50:52 And I think it’s interesting when you start experimenting with it,
1:50:56 how it changes the fluidity of the social spaces
1:50:59 and the way that you interact with them.
1:51:01 But there’s also a commitment.
1:51:02 Like you have to wear the outfit all the time.
1:51:05 I know.
1:51:05 I know.
1:51:06 That’s a big commitment.
1:51:07 Do you feel like that every morning?
1:51:08 No.
1:51:10 I wear, that’s why…
1:51:10 I feel like this is a life commitment.
1:51:14 All I have is suits and black shirt and jeans.
1:51:17 Those are the two outfits.
1:51:18 Yeah.
1:51:19 Well, see, this is the thing though, right?
1:51:20 It simplifies your thought process in the morning.
1:51:22 So like I have other ways I do that.
1:51:24 I park in the same exact parking spot when I go to work
1:51:27 on the fourth floor of a parking garage
1:51:28 because no one ever parks on the fourth floor.
1:51:30 So I don’t have to remember where I park my car.
1:51:32 But I really like aesthetics and playing with them.
1:51:37 So I’m willing to spend part of my cognitive energy
1:51:40 every morning trying to figure out what I want to be that day.
1:51:42 Did you really think about the outfit you’re wearing today?
1:51:46 Yep.
1:51:46 Was there backup options?
1:51:48 Were you going back and forth between some…
1:51:49 Three or four.
1:51:50 But I really like the left.
1:51:51 Were they drastically different?
1:51:53 Yes.
1:51:53 It’s okay.
1:51:56 And then even this one could have been really different
1:51:58 because like, you know, it’s not just the sort of jacket
1:52:01 and the shoes and like, and the hairstyle.
1:52:04 It’s like the jewelry and the accessories.
1:52:05 So like any outfit is a lot of small decisions.
1:52:11 Well, I think your current off is like a lot of shades of yellow.
1:52:15 There’s like a theme.
1:52:16 Yeah.
1:52:17 It’s nice.
1:52:17 It’s really, I’m grateful that you did that.
1:52:21 It’s like it’s it’s it’s own art form.
1:52:23 Yeah.
1:52:23 Yellow is my daughter’s favorite color.
1:52:25 And I never really thought about yellow much,
1:52:27 but she’s been obsessed with yellow.
1:52:28 She’s seven now.
1:52:29 And I don’t know.
1:52:30 I just really love it.
1:52:31 I guess you can pick a color and just make that the constraint.
1:52:35 And then just go with it.
1:52:36 I’m playing with yellow a lot lately.
1:52:37 Like this is not even the most yellow because I have black pants on,
1:52:40 but I have worn outfits that have probably five shades of yellow in them.
1:52:45 Wow.
1:52:45 What what do you think beauty is?
1:52:50 We seem to so underline this idea of playing with aesthetics
1:52:54 is we find certain things beautiful.
1:52:56 Yeah.
1:52:57 What is it that humans find beautiful?
1:53:00 And why do we need to find things beautiful?
1:53:02 Yeah, you know, it’s interesting.
1:53:05 It’s not I’m not I mean, I am attracted to to to style and aesthetics
1:53:11 because I think they’re beautiful,
1:53:12 but it’s much more because I think it’s fun to play with.
1:53:15 And so so I will get to the beauty thing.
1:53:20 But I like I guess I want to just explain a little bit
1:53:23 about my motivation in this space
1:53:24 because it’s really an intellectual thing for me.
1:53:26 And you know, Stuart Brand has this great infographic
1:53:31 about the layers of like human society.
1:53:35 And I think it starts with like the natural sciences
1:53:37 and like physics at the bottom.
1:53:38 And it goes through all these layers and it’s like economics.
1:53:40 And then like fashion is at the top
1:53:41 is like the fastest moving part of human culture.
1:53:45 And I think I really like that because it’s so dynamic
1:53:48 and so short and it’s temporal longevity.
1:53:51 Contrasted with like studying the laws of physics,
1:53:55 which are like, you know, like the deep structure
1:53:57 reality that I feel like I like bridging those scales
1:53:59 tells me much more about the structure of the world that I live in.
1:54:04 That said, there’s some kinds of fashions,
1:54:06 like a dude in a black suit with a black tie.
1:54:10 It seems to be less dynamic.
1:54:14 Yeah.
1:54:14 It seems to persist through time.
1:54:16 Are you embodying this?
1:54:17 Yeah, I think so.
1:54:18 I think I think it just.
1:54:21 I’d like to see you wear yellow.
1:54:23 I wouldn’t even know what to do with myself.
1:54:27 I would freak out.
1:54:28 I wouldn’t know how to act in the world.
1:54:29 You wouldn’t know how to be you.
1:54:31 Yeah.
1:54:31 I know this is amazing though, isn’t it?
1:54:33 Amazing. Like you have the choice to do it.
1:54:35 But what are my favorite, just on the question of beauty,
1:54:38 one of my favorite fashion designers of all time
1:54:40 is Alexander McQueen.
1:54:41 And he was really phenomenal.
1:54:45 But like his early, and actually I kind of used
1:54:49 like what happened to him in the fashion industry
1:54:50 is a coping mechanism with our paper when,
1:54:52 like the nature paper in the fall,
1:54:55 when everyone was saying it was controversial
1:54:57 and how terrible that like, you know, like,
1:54:58 but controversial is good, right?
1:54:59 But like Alexander McQueen, you know,
1:55:01 first came out with his fashion lines.
1:55:03 He was mixing horror and beauty.
1:55:05 And people were horrified.
1:55:07 It was so controversial.
1:55:09 Like they, like it was macabre.
1:55:10 He had like, you know, like it looked like
1:55:12 there were blood on the models and like.
1:55:14 That’s beautiful.
1:55:16 We just look into pictures here.
1:55:17 Yeah. No, I mean, his stuff is amazing.
1:55:20 His first like runway line, I think was called Neilism.
1:55:25 I don’t know if you could find it.
1:55:27 You know, I mean, he was really dramatic.
1:55:30 He, he carried a lot of trauma with him.
1:55:32 There you go. That’s yeah.
1:55:33 Yeah. But he changed the fashion industry.
1:55:38 His stuff became very popular.
1:55:40 That’s a good offer to show up to a party.
1:55:42 Right. Right.
1:55:44 But this gets at the question, like,
1:55:45 is that horrific or is it beautiful?
1:55:48 And I think, you know, he, he had a traumatic,
1:55:52 he ended up committing suicide.
1:55:56 And actually he left his death note on the descent of man.
1:55:59 So he was, he was a really deep person.
1:56:02 So I mean, great fashion certainly has that kind of depth to it.
1:56:05 Yeah, it sure does.
1:56:07 So I think it’s the intellectual pursuit, right?
1:56:09 Like it’s not, so this is like very highly intellectual.
1:56:12 And I think it’s a lot like how I play with language
1:56:14 is the same way that I play with fashion
1:56:16 or the same way that I play with ideas and theoretical physics.
1:56:19 Like there’s always this space
1:56:20 that you can just push things just enough.
1:56:23 So they’re like, they look like something,
1:56:25 someone thinks is familiar, but they’re not familiar.
1:56:29 And yeah, and I think that’s really cool.
1:56:31 It seems like beauty doesn’t have much function, right?
1:56:35 But, but it seems to also have
1:56:37 a lot of influence on the way we collaborate with each other.
1:56:42 It has tons of function.
1:56:43 What do you mean it doesn’t have function?
1:56:44 I guess sexual selection incorporates beauty somehow.
1:56:47 But why?
1:56:48 Because beauty is a sign of health or something.
1:56:51 I don’t even.
1:56:52 Oh, evolutionarily, maybe.
1:56:55 But then beauty becomes a signal of other things, right?
1:56:57 So it’s really not like, and then beauty becomes
1:57:00 an adaptive trait.
1:57:01 So it can change with different species.
1:57:03 Like, you know, maybe some people, some species would think,
1:57:05 well, you thought the frog having babies come out of its back
1:57:08 was beautiful and I thought it was grotesque.
1:57:10 Like there’s not a universal definition of what’s beautiful.
1:57:13 It is something that is dependent on your history
1:57:17 and how you interact with the world.
1:57:19 And I guess what I like about beauty,
1:57:23 like any other concept is when you turn it on its head.
1:57:25 So, you know, maybe the traditional conception of, you know,
1:57:31 why women wear makeup and they dress certain ways
1:57:35 is because they want to look beautiful and pleasing to people.
1:57:40 And I just like to do it because it’s a confidence thing.
1:57:44 It’s about embodying the person that I want to be
1:57:48 and about owning that person.
1:57:52 And then the way that people interact with that person
1:57:54 is very different than if I didn’t have that.
1:57:56 Like if I wasn’t using that attribute as part of…
1:57:59 And obviously that’s influenced by the society I live
1:58:03 and like what’s aesthetically pleasing things.
1:58:05 But it’s interesting to be able to turn that around
1:58:06 and not have it necessarily be about the aesthetics
1:58:09 but about the power dynamics that the aesthetics create.
1:58:11 But you’re saying there’s some function to beauty
1:58:14 in that way, in the way you’re describing
1:58:15 and the dynamic it creates in the social interaction.
1:58:18 Well, the point is you’re saying it’s an adaptive trait
1:58:20 for like sexual selection or something.
1:58:22 And I’m saying that the adaptation that beauty confers
1:58:25 is far richer than that.
1:58:27 And some of the adaptation is about social hierarchy
1:58:30 and social mobility and just plain social dynamics.
1:58:34 Like why do some people dress golf?
1:58:36 It’s because they identify with a community
1:58:38 and a culture associated with that and they get…
1:58:40 And that’s a beautiful aesthetic.
1:58:43 It’s a different aesthetic.
1:58:44 Some people don’t like it.
1:58:45 So it has the same richness as the language?
1:58:49 Yes.
1:58:50 It’s the same kind of…
1:58:51 Yes. And I think too few people think about the way that they…
1:58:57 The aesthetics they build for themselves
1:58:59 in the morning and how they carry it in the world
1:59:01 and the way that other people interact with that
1:59:03 because they put clothes on
1:59:05 and they don’t think about clothes as carrying function.
1:59:07 Let’s jump from beauty to language.
1:59:10 There’s so many ways to explore the topic of language.
1:59:14 You called it…
1:59:15 You said that language is…
1:59:18 Parts of language or language in itself
1:59:20 and the mechanism of language is a kind of living life form.
1:59:23 You’ve tweeted a lot about this in all kinds of poetic ways.
1:59:28 Let’s talk about the computation aspect of it.
1:59:30 You tweeted, “The world is not a computation
1:59:35 but computation is our best current language
1:59:37 for understanding the world.”
1:59:38 It is important we recognize this
1:59:40 so we can start to see the structure of our future languages
1:59:43 that will allow us to see deeper than the computation allows us.
1:59:48 So what’s the use of language in helping us understand
1:59:51 and make sense of the world?
1:59:51 I think one thing that I feel like I notice much more viscerally
1:59:57 than I feel like I hear other people describe
2:00:00 is that the representations in our mind
2:00:05 and the way that we use language are not the things like…
2:00:10 Actually, I mean, this is an important point
2:00:14 going back to what Gertl did
2:00:15 but also this idea of signs and symbols
2:00:17 and all kinds of ways of separating them.
2:00:19 There’s like the word, right?
2:00:21 And then there’s like what the word means about the world
2:00:24 and we often confuse those things.
2:00:26 And what I feel very viscerally…
2:00:31 I almost sometimes think I have some kind of like
2:00:33 synesthesia for language or something
2:00:35 and I just like don’t interact with it
2:00:36 like the way that other people do.
2:00:37 But for me, words are objects
2:00:40 and the objects are not the things that they describe.
2:00:42 They have like a different ontology to them.
2:00:45 Like they’re physical things and they carry causation
2:00:49 and they can create meaning.
2:00:50 But they’re not what we think they are.
2:00:56 And also like the internal representations in our mind
2:00:59 like the things I’m seeing about this room
2:01:01 are probably like there’s small projection
2:01:03 of the things that are actually in this room.
2:01:06 And I think we have such a difficult time moving past
2:01:09 the way that we build representations in the mind
2:01:12 and the way that we structure our language
2:01:13 to realize that those are approximations
2:01:15 to what’s out there and they’re fluid
2:01:17 and we can play around with them
2:01:18 and we can see deeper structure underneath them,
2:01:20 that I think like we’re missing a lot.
2:01:23 Yeah, but also the life of the mind is in some ways richer
2:01:27 than the physical reality.
2:01:29 Sure.
2:01:29 What’s going on in your mind,
2:01:30 it might be a projection actually here,
2:01:34 but there’s also all kinds of other stuff going on there.
2:01:38 Yeah, for sure.
2:01:39 I love this essay by Juan Correa
2:01:41 about like mathematical creativity
2:01:43 where he talks about this sort of like frothing
2:01:45 of all these things and then like somehow
2:01:46 you build theorems on top of it and they become kind of concrete.
2:01:49 But like, and I also think about this with language,
2:01:52 it’s like there’s a lot of stuff happening in your mind,
2:01:54 but you have to compress it in this few sets of words
2:01:57 to try to convey it to someone.
2:01:59 So it’s a compactification of the space.
2:02:02 And it’s not a very efficient one.
2:02:05 And I think just recognizing that there’s a lot
2:02:09 that’s happening behind language is really important.
2:02:11 And I think this is one of the great things
2:02:14 about the existential trauma of large language models,
2:02:17 I think is the recognition that language
2:02:20 is not the only thing required.
2:02:21 Like there’s something underneath it, not by everybody.
2:02:26 Can you just speak to the feeling you have
2:02:32 when you think about words?
2:02:33 So is there like, what’s the magic of words to you?
2:02:36 Is it like, do you feel it is almost sometimes
2:02:39 feels like you’re playing with it?
2:02:42 Yeah, I was just going to say it’s like a playground.
2:02:44 But you’re almost like, I think one of the things
2:02:47 you enjoy, maybe I’m projecting, is deviating,
2:02:51 like using words in ways that not everyone uses them.
2:02:54 Like slightly sort of deviating from the norm a little bit.
2:02:58 I love doing that in everything I do,
2:03:00 but especially with language.
2:03:01 But not so far, that doesn’t make sense.
2:03:04 Exactly.
2:03:05 So you’re always like tethered to reality, to the norm,
2:03:10 but like are playing with it,
2:03:11 like basically fucking with people’s minds a little bit.
2:03:15 I mean, like, you know, and in so doing,
2:03:18 creating a different perspective on the thing
2:03:21 that’s been previously explored in a different way.
2:03:24 Yeah, it’s literally my favorite thing to do.
2:03:26 Yeah, like use words as one way to make people think.
2:03:31 Yeah, so I, you know, a lot of my sort of like what happens
2:03:36 in my mind when I’m thinking about ideas
2:03:39 is I’ve been presented with this information
2:03:41 about how people think about things.
2:03:42 And I try to go around to different communities
2:03:46 and hear the ways that different, whether it’s like,
2:03:49 you know, hanging out with a bunch of artists
2:03:50 or philosophers or scientists thinking about things.
2:03:54 Like they all think about it different ways.
2:03:55 And then I just try to figure out like,
2:03:59 how do you take the structure of the way
2:04:00 that we’re talking about it and turn it slightly?
2:04:04 So you have all the same pieces that everybody sees are there,
2:04:08 but the description that you’ve come up with
2:04:10 seems totally different.
2:04:11 So they can understand that there’s,
2:04:12 like they understand the pattern you’re describing,
2:04:15 but they never heard the structure underlying it describe
2:04:18 the way that you describe it.
2:04:19 Is there words or terms you remember that
2:04:26 disturbed people the most,
2:04:28 maybe the positive sense of disturbed?
2:04:30 There’s assembly theory, I suppose is one.
2:04:33 Yeah. I mean, the first couple sentences
2:04:36 of that paper disturbed people a lot.
2:04:38 And I think they were really carefully constructed
2:04:40 in exactly this kind of way.
2:04:41 What was that? Let me look it up.
2:04:43 Oh, it was really fun.
2:04:43 But I think it’s interesting because I do, you know,
2:04:50 sometimes I’m very upfront about it.
2:04:52 I say I’m going to use the same word
2:04:53 in probably six different ways.
2:04:55 In a lecture and I will.
2:04:59 You’re right. Scientists have grappled
2:05:01 with reconciling biological evolution
2:05:03 with immutable laws of the universe defined by physics.
2:05:06 These laws underpin life’s origin, evolution, and the development
2:05:12 of human culture.
2:05:14 Well, he was, I think your love for words runs deeper than these.
2:05:19 Yeah, for sure.
2:05:20 I mean, this is part of the sort of brilliant thing
2:05:24 about our collaboration is, you know, complementary skillsets.
2:05:30 So I love playing with the abstract space of language.
2:05:34 And it’s a really interesting playground
2:05:37 when I’m working with Lee because he thinks
2:05:41 at a much deeper level of abstraction
2:05:43 than can be expressed by language.
2:05:45 And the ideas we work on are hard to talk about for that reason.
2:05:49 What do you think about computation as a language?
2:05:52 I think it’s a very poor language.
2:05:54 A lot of people think it’s a really great one,
2:05:55 but I think it has some nice properties.
2:05:57 But I think that the feature of it that, you know,
2:06:01 is compelling is this kind of idea of universality
2:06:04 that like you can, if you have a language,
2:06:08 you can describe things in any other language.
2:06:10 Well, for me, one of the people who kind of revealed
2:06:13 the expressive power of computation,
2:06:16 aside from Alan Turing,
2:06:18 is Stephen Wolfram through all the explorations
2:06:20 of like cellular automata type of objects
2:06:22 that he did in a new kind of science.
2:06:25 And afterwards, so what would he get from that?
2:06:28 The kind of computational worlds that are revealed
2:06:34 through even something as simple as cellular automata.
2:06:37 It seems like that’s a really nice way to explore languages
2:06:41 that are far outside our human languages
2:06:46 and do so rigorously and understand
2:06:49 how those kinds of complex systems can interact
2:06:54 with each other, can emerge, all that kind of stuff.
2:06:56 I don’t think that they’re outside our human languages.
2:07:01 I think they define the boundary
2:07:03 of the space of human languages.
2:07:05 They allow us to explore things within that space,
2:07:08 which is also fantastic.
2:07:09 But I think there is a set of ideas that takes,
2:07:11 and Stephen Wolfram has worked on this quite a lot,
2:07:16 and contributed very significantly to it.
2:07:18 And I really like some of the stuff
2:07:21 that Stephen’s doing with his physics project,
2:07:23 but don’t agree with a lot of the foundations of it.
2:07:25 But I think the space is really fun that he’s exploring.
2:07:28 There’s this assumption that computation
2:07:32 is at the base of reality.
2:07:33 And I kind of see it at the top of reality,
2:07:37 not at the base,
2:07:39 because I think computation was built by our biosphere.
2:07:42 It’s something that happened
2:07:43 after many billion years of evolution.
2:07:46 And it doesn’t happen in every physical object.
2:07:49 It only happens in some of them.
2:07:51 And I think one of the reasons
2:07:53 that we feel like the universe is computational
2:07:57 is because it’s so easy for us as things
2:08:01 that have the theory of computation in our minds.
2:08:06 And actually, in some sense,
2:08:07 it might be related to the functioning of our minds
2:08:10 and how we build languages to describe the world
2:08:13 and sets of relations to describe the world.
2:08:15 But it’s easy for us to go out into the world
2:08:20 and build computers.
2:08:22 And then we mistake our ability to do that
2:08:25 with assuming that the world is computational.
2:08:27 And I’ll give you a really simple example.
2:08:30 This one came from John Conway.
2:08:32 I one time had a conversation with him,
2:08:34 which was really delightful.
2:08:36 He was really fun.
2:08:37 But he was pointing out that if you string lights in a barn,
2:08:44 you can program them to have your favorite one-dimensional C.A.
2:08:49 And you might even be able to make them be capable
2:08:53 of universal computation.
2:08:54 Is universal computation a feature of the string lights?
2:08:58 Well, no.
2:08:59 No. It’s probably not.
2:09:01 It’s a feature of the fact that you, as a programmer,
2:09:04 had a theory that you could embed
2:09:06 in the physical architecture of the string lights.
2:09:08 Now, what happens, though,
2:09:10 is we get confused by this kind of distinction
2:09:12 between us as agents in the world
2:09:14 that actually can transfer things
2:09:16 that life does onto other physical substrates
2:09:19 with what the world is.
2:09:20 And so, for example, you’ll see people
2:09:23 doing– studying the mathematics of chemical reaction networks
2:09:27 and saying, well, chemistry is turning universal
2:09:30 or studying the laws of physics
2:09:31 and saying the laws of physics are turning universal.
2:09:34 But anytime that you want to do that,
2:09:36 you always have to prepare an initial state.
2:09:38 You have to constrain the rule space.
2:09:41 And then you have to actually be able to demonstrate
2:09:44 the properties of computation.
2:09:46 And all of that requires an agent or a designer
2:09:48 to be able to do that.
2:09:49 But it gives you an intuition.
2:09:52 If you look at a 1D or 2D cellular automata,
2:09:55 it gives you– it allows you to build an intuition
2:09:59 of how you can have complexity emerge
2:10:01 from very simple beginnings.
2:10:03 Very simple initial conditions.
2:10:04 I think that’s the intuition that people have derived from it.
2:10:07 The intuition I get from cellular automata
2:10:11 is that the flat space of an initial condition
2:10:13 in a fixed dynamical law is not rich enough
2:10:15 to describe an open-ended generation process.
2:10:18 And so the way I see cellular automata
2:10:20 is they’re embedded slices in a much larger causal structure.
2:10:23 And if you want to look at a deterministic slice
2:10:25 of that causal structure, you might be able to extract
2:10:27 a set of consistent rules that you might call a cellular automata,
2:10:30 but you could embed them as much larger space.
2:10:33 That’s not dynamical and is about the causal structure
2:10:36 and relations between all of those computations.
2:10:38 And that would be the space cellular automata live in.
2:10:41 And I think that’s the space that Stephen is talking about
2:10:45 when he talks about his RULIAD
2:10:46 and these hypergraphs of all these possible computations.
2:10:49 But I wouldn’t take that as my base reality
2:10:52 because I think, again, computation itself,
2:10:54 this abstract property computation,
2:10:56 is not at the base of reality.
2:10:58 So can we just link on that RULIAD this–
2:11:01 Yeah.
2:11:01 One RULIAD to rule them all.
2:11:04 Yeah.
2:11:05 So this is part of Wolfram physics project.
2:11:09 It’s what he calls the entangled limit
2:11:12 of everything that is computationally possible.
2:11:14 So what’s your problem with the RULIAD?
2:11:18 Well, it’s interesting.
2:11:20 So Stephen came to a workshop we had in the Beyond Center
2:11:23 in the fall.
2:11:24 And the workshop theme was mathematics.
2:11:26 Is it evolved or eternal?
2:11:28 And he gave a talk about the RULIAD.
2:11:30 And he was talking about how a lot of the things
2:11:33 that we talk about in the Beyond Center,
2:11:35 like, does reality have a bottom?
2:11:37 If it has a bottom, what is it?
2:11:38 You know, like–
2:11:40 I need to go–
2:11:42 We’ll have you to one sometime.
2:11:44 No, this is great.
2:11:45 Does reality have a bottom?
2:11:48 Yeah.
2:11:48 So we had one that was called infinite turtles or ground
2:11:52 truth.
2:11:53 And it was really just about this issue.
2:11:56 But the thing that was interesting,
2:11:57 I think Stephen was trying to make the argument
2:12:00 that fundamental particles aren’t fundamental,
2:12:03 gravitation is not fundamental.
2:12:04 These are just turtles.
2:12:08 And computation is fundamental.
2:12:10 And I remember pointing out to him,
2:12:12 I was like, well, computation is your turtle.
2:12:14 And I think it’s a weird turtle to have.
2:12:17 First of all, isn’t it OK to have a turtle?
2:12:20 It’s totally fine to have a turtle.
2:12:22 Everyone has a turtle.
2:12:23 You can’t build a theory without a turtle.
2:12:25 It depends on the problem you want to describe.
2:12:29 And actually, the reason I can’t get behind
2:12:32 Stephen’s ontology is I don’t know what question
2:12:35 he’s trying to answer.
2:12:37 And without a question to answer,
2:12:38 I don’t understand why you’re building a theory of reality.
2:12:40 And the question you’re trying to answer is–
2:12:42 What life is.
2:12:44 What life is, which another simpler way of phrasing that
2:12:48 is how did life originate?
2:12:50 Well, I started working on the origin of life.
2:12:52 And I think what my challenge was there
2:12:56 was no one knew what life was.
2:12:57 And so you can’t really talk about the origination
2:12:59 of something if you don’t know what it is.
2:13:01 And so the way I would approach it
2:13:04 is if you want to understand what life is,
2:13:06 then proving that physics is solving the origin of life.
2:13:10 So there’s the theory of what life is,
2:13:13 but there’s the actual demonstration
2:13:15 that that theory is an accurate description
2:13:17 of the phenomena you aim to describe.
2:13:18 So again, they’re the same problem.
2:13:20 It’s not like I can decouple origin of life from what life is.
2:13:23 It’s like that is the problem.
2:13:26 And the point I guess I’m making about having a question
2:13:30 is no matter what slice of reality you take,
2:13:34 what regularity of nature you’re going to try to describe,
2:13:36 there will be an abstraction
2:13:40 that unifies that structure of reality, hopefully.
2:13:43 And that will have a fundamental layer to it,
2:13:49 because you have to explain something
2:13:52 in terms of something else.
2:13:53 But so if I want to explain life, for example,
2:13:56 then my fundamental description of nature
2:13:58 has to be something I think that has to do
2:14:00 with time being fundamental.
2:14:01 But if I wanted to describe,
2:14:03 I don’t know, the sort of interactions of matter and light,
2:14:11 I have elementary particles be fundamental.
2:14:13 If I want to describe electricity and magnetism in the 1800s,
2:14:17 I have to have waves be fundamental, right?
2:14:20 So like you earn quantum mechanics,
2:14:23 like it’s a wave function that’s fundamental
2:14:25 because that’s the sort of explanatory paradigm
2:14:28 of your theory.
2:14:28 So I guess I don’t know what problem
2:14:35 saying computation is fundamental solves.
2:14:39 Doesn’t he want to understand
2:14:42 how does the basic quantum mechanics
2:14:45 and general relativity emerge and cut us time?
2:14:49 Right, so I think–
2:14:50 But then that doesn’t really answer an important question for us.
2:14:52 Well, I think the issue is general relativity
2:14:55 and quantum mechanics are expressed in mathematical languages.
2:14:58 And then computation is a mathematical language.
2:15:02 So you’re basically saying that maybe there’s
2:15:04 a more universal mathematical language
2:15:06 for describing theories of physics that we already know.
2:15:08 That’s an important question,
2:15:09 and I do think that’s what Stephen’s trying to do and do well.
2:15:11 But then the question becomes,
2:15:15 does that formulation of a more universal language
2:15:18 for describing the laws of physics that we know now
2:15:22 tell us anything new about the nature of reality?
2:15:24 Or is it a language?
2:15:26 And to you, languages can be fundamental.
2:15:31 The language itself is never the fundamental thing.
2:15:35 It’s whatever it’s describing.
2:15:37 So one of the possible titles you were thinking about
2:15:39 originally for the book is the hard problem of life,
2:15:43 sort of reminiscent of the hard problem of consciousness.
2:15:47 So you’re saying that assembly theory
2:15:49 is supposed to be answering the question
2:15:51 about what is life.
2:15:52 So let’s go to the other hard problems.
2:15:55 You also say that the easiest of the hard problems
2:15:58 is the hard problem of life.
2:16:01 So what do you think is the nature of intelligence and consciousness?
2:16:09 We think something like assembly theory can help us understand that.
2:16:19 I think if assembly theory is an accurate depiction of the physics of life,
2:16:27 it should shed a lot of light on those problems.
2:16:31 And in fact, I sometimes wonder if the problems of consciousness
2:16:34 and intelligence are at all different than the problem of life, generally.
2:16:38 And I’m of two minds of it, but I in general try to,
2:16:46 you know, like the process of my thinking
2:16:49 is trying to regularize everything into one theory.
2:16:51 So pretty much every direction I have is like,
2:16:54 “Oh, how do I fold that into?”
2:16:56 And like, so I’m just building this giant abstraction
2:16:58 that’s basically trying to take every piece of data I’ve ever gotten
2:17:01 in my brain into a theory of what life is.
2:17:04 And consciousness and intelligence are obviously
2:17:08 some of the most interesting things that life has manifest.
2:17:11 And so I think they’re very telling about some of the deeper features
2:17:16 about the nature of life.
2:17:18 This seems like they’re all flavors of the same thing.
2:17:22 But it’s interesting to wonder at which stage
2:17:24 that’s something that we would recognize as life
2:17:28 in a sort of canonical, silly human way
2:17:32 and something that we would recognize as intelligence.
2:17:35 At which stage does that emerge?
2:17:37 Like at which assembly index does that emerge?
2:17:39 And at which assembly index is consciousness?
2:17:42 Something that we would canonically recognize as consciousness?
2:17:45 Is this the use, like this use of flavors the same as you meant
2:17:48 when you were talking about flavors of alien life?
2:17:51 Yeah, sure. Yeah.
2:17:53 I mean, it’s the same as the flavors of ice cream
2:17:56 and the flavors of fashion.
2:17:57 Yeah, like, but we were talking about in terms of colors
2:18:00 and like very nondescript.
2:18:01 But the way that you just talked about flavors now
2:18:03 was more in like the space of consciousness and intelligence.
2:18:05 It was kind of like much more specific.
2:18:07 It’d be nice if there’s a formal way of expressing.
2:18:11 Quantifying flavors.
2:18:13 Quantifying flavors.
2:18:14 It seems like I would order life consciousness intelligence,
2:18:21 probably, as like the order in which things emerge.
2:18:25 And they’re all just the same.
2:18:27 We’re using the word life differently here.
2:18:32 I mean, life sort of when I’m talking about what is
2:18:35 a living versus non-living thing at a bar with a person,
2:18:38 I’m already like four or five drinks in that kind of thing.
2:18:42 Like we’re not we’re not being too philosophical.
2:18:46 Like there’s a thing that moves and here’s the thing that doesn’t move.
2:18:49 And but maybe consciousness precedes that.
2:18:53 It’s a weird dance there.
2:18:57 Is life precede consciousness or consciousness precede life?
2:19:03 And I think that understanding of what life is
2:19:07 and the way you’re doing will help us disentangle that.
2:19:10 Depending on what you want to explain, as I was saying before,
2:19:13 you have to assume something’s fundamental.
2:19:15 And so because people can’t explain consciousness,
2:19:17 there’s a temptation for some people to want to take consciousness
2:19:21 as fundamental and assume everything else is derived out of that.
2:19:24 And then you get some people that want to assume consciousness preceded life.
2:19:29 And I don’t I don’t find either of those views particularly illuminating.
2:19:33 I think because I don’t I don’t want to assume a feminology before I explain a thing.
2:19:40 And so what I’ve tried really hard to do is is not assume
2:19:44 that I think life is anything except hold on to sort of the patterns and structures
2:19:49 that seem to be the sort of consistent ways that we talk about this thing
2:19:52 and then try to build a physics that describes that.
2:19:55 And I think that’s a really different approach than saying,
2:19:58 you know, consciousness is this thing, you know, we all feel and experience about things.
2:20:03 I would want to understand the regularities associated with that
2:20:07 and build a deeper structure underneath that and build into it.
2:20:10 I wouldn’t want to assume that thing and that I understand that thing,
2:20:14 which is usually how I see people talk about it.
2:20:16 The difference between life and consciousness, which which comes first.
2:20:21 Yeah, so I think if you’re thinking about this sort of thinking about living things
2:20:28 as these giant causal structures or these objects that are deep in time
2:20:32 or whatever language we end up using to describe it.
2:20:35 It seems to me that consciousness is about the fact that we have a conscious experience
2:20:46 is because we are these temporally extended objects.
2:20:49 So consciousness and the abstraction that we have in our minds
2:20:53 is actually a manifestation of all the time that’s rolled up in us.
2:20:56 And it’s just because we’re so huge that we have this very large inner space
2:21:00 that we’re experiencing.
2:21:01 That’s not and it’s also separated off from the rest of the world
2:21:04 because we’re the separate thread in time.
2:21:06 And so our consciousness is not exactly shared with anything else
2:21:10 because nothing else occupies the same part of time that we occupy.
2:21:15 But I can understand something about you maybe being conscious
2:21:19 because you and I didn’t separate that far in the past
2:21:22 in terms of our causal histories.
2:21:26 So in some sense, we can even share experiences with each other through language
2:21:30 because of that sort of overlap in our structure.
2:21:33 Well, then if consciousness is merely temporal separateness,
2:21:38 then that comes before life.
2:21:40 It’s not merely temporal separateness.
2:21:43 It’s about the depth in that time.
2:21:45 So it’s the reason that my conscious experience is not the same as yours
2:21:49 is because we’re separated in time.
2:21:50 The fact that I have a conscious experience is because I’m an object
2:21:53 that’s super deep in time.
2:21:54 So I’m huge in time and that means that there’s a lot there that I am basically
2:22:00 in some sense a universe onto myself because my structure is so large
2:22:04 relative to the amount of space that I occupy.
2:22:06 But it feels like that’s possible to do before you get anything like bacteria.
2:22:13 I think there’s a horizon and I don’t know how to articulate this yet.
2:22:17 It’s a little bit like the horizon at the origin of life
2:22:19 where the space inside a particular structure becomes so large
2:22:24 that it has some access to a space that doesn’t feel as physical.
2:22:31 It’s almost like this idea of counterfactuals.
2:22:33 So I think the past history of your horizon is just much larger than can be encompassed
2:22:42 in a small configuration of matter.
2:22:44 So you can pull this stuff into existence.
2:22:47 This property is maybe a continuous property,
2:22:50 but there’s something really different about human level physical systems
2:22:56 and human level ability to understand reality.
2:23:00 I really love David Deutsch’s conception of universal explainers
2:23:04 and that’s related to the theory of universal computation.
2:23:09 And I think there’s some transition that happens there.
2:23:13 But maybe to describe that a little bit better,
2:23:17 what I can also say is what intelligence is in this framework.
2:23:20 So you have these objects that are large in time.
2:23:24 They were selected to exist by constraining the possible space of objects to this particular,
2:23:31 like all of the matters, funneled into this particular configuration of object over time.
2:23:37 And so these objects arise through selection.
2:23:40 But the more selection that you have embedded in you,
2:23:43 the more possible selection you have on your future.
2:23:45 And so selection and evolution we usually think about in the past sense,
2:23:52 where selection happened in the past.
2:23:54 But objects that are high density configurations of matter that have a lot of selection in them
2:24:01 are also selecting agents in the universe.
2:24:04 So they actually embody the physics of selection and they can select on possible futures.
2:24:08 And I guess what I’m saying with respect to consciousness and the experience we have
2:24:13 is that there’s something very deep about that structure
2:24:16 and the nature of how we exist in that structure
2:24:19 that has to do with how we’re navigating that space
2:24:23 and how we generate that space and how we continue to persist in that space.
2:24:28 Is there shortcuts we can take to artificially engineering living organisms, artificial life,
2:24:36 artificial consciousness, artificial intelligence?
2:24:39 So maybe just looking pragmatically at the LLMs we have now.
2:24:46 Do you think those can exhibit qualities of life, qualities of consciousness,
2:24:53 qualities of intelligence in the way we think of intelligence?
2:24:57 I mean, I think they already do, but not in the way I hear popularly discussed.
2:25:01 So there are obviously signatures of intelligence
2:25:04 and a part of an ecosystem of intelligence systems.
2:25:11 But I don’t know that individually, I would assign all the properties to them that people have.
2:25:18 It’s a little like, so we talked about the history of eyes before
2:25:22 and how eyes scaled up into technological forms.
2:25:25 And language has also had a really interesting history
2:25:29 and got much more interesting, I think, once we started writing it down
2:25:32 and then inventing books and things.
2:25:35 But every time that we started storing language in a new way,
2:25:41 we were kind of existentially traumatized by it.
2:25:45 So the idea of written language was traumatic
2:25:48 because it seemed like the dead were speaking to us,
2:25:50 even though they were deceased and books were traumatic,
2:25:52 because suddenly there were lots of copies of this information available to everyone
2:25:58 and it was going to somehow dilute it.
2:26:01 And large language models are kind of interesting
2:26:04 because they don’t feel as static, they’re very dynamic.
2:26:07 But if you think about language in the way I was describing before,
2:26:09 as language is this very large in time structure
2:26:12 and before it had been something that was distributed over human brains as a dynamic structure,
2:26:18 and occasionally we store components of that very large dynamic structure
2:26:23 in books or in written language,
2:26:25 now we can actually store the dynamics of that structure
2:26:29 in a physical artifact, which is a large language model.
2:26:32 And so I think about it almost like the evolution of genomes in some sense,
2:26:37 where there might have been really primitive genes in the first living things
2:26:41 and they didn’t store a lot of information or they were really messy.
2:26:45 And then by the time you get to the U.K. or XL,
2:26:47 you have this really dynamic genetic architecture that’s read-writeable
2:26:50 and has all of these different properties.
2:26:53 And I think large language models are kind of like the genetic system for language
2:26:58 in some sense, where it’s allowing a sort of archiving that’s highly dynamic.
2:27:05 And I think it’s very paradoxical to us because obviously in human history,
2:27:10 we haven’t been used to conversing with anything that’s not human.
2:27:15 But now we can converse basically with a crystallization of human language in a computer.
2:27:23 That’s a highly dynamic crystal because it’s a crystallization in time of this
2:27:28 massive abstract structure that’s evolved over human history
2:27:31 and is now put into a small device.
2:27:34 I think crystallization kind of implies a limit on its capabilities.
2:27:40 I mean it very purposefully because a particular instantiation of a language model
2:27:46 trained on a particular dataset becomes a crystal of the language at that time it was
2:27:50 trained, but obviously we’re iterating with the technology and evolving it.
2:27:53 I guess the question is when you crystallize it, when you compress it, when you archive it,
2:27:58 you’re archiving some slice of the collective intelligence of the human species.
2:28:04 That’s right.
2:28:05 And the question is how powerful is that?
2:28:09 Right, it’s a societal level technology, right?
2:28:11 We’ve actually put collective intelligence in a box.
2:28:14 Yeah, I mean how much smarter is the collective intelligence of humans versus a single human?
2:28:20 And that’s the question of AGI versus human level intelligence,
2:28:27 superhuman level intelligence versus human level intelligence.
2:28:30 Like how much smarter can this thing, when done well, when we solve a lot of the
2:28:36 complexity, computation complexities, maybe there’s some data complexities and how to
2:28:41 really archive this thing, crystallize this thing really well.
2:28:44 How powerful is this thing going to be?
2:28:46 I actually, I don’t like the sort of language we use around that.
2:28:53 And I think the language really matters.
2:28:54 So I don’t know how to talk about how much smarter one human is than another, right?
2:29:00 Like usually we talk about abilities or particular talents someone has.
2:29:07 And going back to David Rich’s idea of universal explainers,
2:29:14 it like adopting the view that we’re the first kinds of structures our biosphere has built
2:29:22 that can understand the rest of reality.
2:29:25 We have this universal comprehension capability.
2:29:29 He makes an argument that basically we’re the first things that actually are capable
2:29:34 of understanding anything.
2:29:35 It doesn’t matter.
2:29:36 It doesn’t mean an individual understands everything, but we have that capability.
2:29:41 And so there’s not a difference between that and what people talk about with AGI.
2:29:45 In some sense, AGI is a universal explainer.
2:29:48 But it might be that a computer is much more efficient at doing,
2:29:55 I don’t know, prime factorization or something than a human is.
2:29:59 But it doesn’t mean that it’s necessarily smarter or has a broader reach of the kind
2:30:05 of things that can understand than a human does.
2:30:08 And so I think we really have to think about, is it a level shift?
2:30:12 Or is it we’re enhancing certain kinds of capabilities humans have in the same way
2:30:18 that we can’t enhanced eyesight by making telescopes and microscopes?
2:30:22 Are we enhancing capabilities we have into technologies and the entire global ecosystem
2:30:27 is getting more intelligent?
2:30:29 Or is it really that we’re building some super machine in a box that’s going to be
2:30:32 smart and kill everybody?
2:30:33 It’s not even a science fiction narrative.
2:30:37 It’s a bad science fiction narrative.
2:30:39 I just don’t think it’s actually accurate to any of the technologies we’re building
2:30:42 or the way that we should be describing them.
2:30:43 It’s not even how we should be describing ourselves.
2:30:46 So the benevolent stories is a benevolent system that’s able to transform our economy,
2:30:52 our way of life by just 10xing the GDP.
2:30:58 Well, these are human questions, right?
2:31:00 I don’t think they’re necessarily questions that we’re going to outsource to an artificial
2:31:05 intelligence.
2:31:06 I think what is happening and will continue to happen is there’s a co-evolution between
2:31:11 humans and technology that’s happening and we’re co-existing in this ecosystem right now
2:31:18 and we’re maintaining a lot of the balance.
2:31:20 And for the balance to shift to the technology would require some very bad human actors,
2:31:25 which is a real risk, or some sort of, I don’t know, some sort of dynamic that favors,
2:31:36 like I just don’t know how that plays out without human agency actually trying to
2:31:41 put it in that direction.
2:31:42 It could also be how rapid the rate is.
2:31:45 The rapid rate is scary.
2:31:47 So I think the things that are terrifying are the ideas of deep fakes or all the kinds
2:31:57 of issues that become legal issues about artificial intelligence technologies
2:32:03 and using them to control weapons or using them for child pornography or faking out that someone’s
2:32:14 loved one was kidnapped or killed and there’s all kinds of things that are super scary in
2:32:21 this landscape and all kinds of new legislation needs to be built and all kinds of guardrails
2:32:27 on the technology to make sure that people don’t abuse it and need to be built.
2:32:30 And that needs to happen.
2:32:32 And I think one function of sort of the artificial intelligence doomsday sort of part of our culture
2:32:41 right now is it’s sort of our immune response to knowing that’s coming.
2:32:45 And we’re overscaring ourselves so we try to act more quickly, which is good.
2:32:50 But I just, you know, it’s about the words that we use versus the actual things happening
2:32:58 behind the words.
2:32:59 I think one thing that’s good is when people are talking about things different ways,
2:33:03 it makes us think about them.
2:33:04 And also when things are existentially threatening, we want to pay attention to those.
2:33:09 But the ways that they’re existentially threatening and the ways that we’re experiencing
2:33:13 existential trauma, I don’t think that we’re really going to understand for another century or two,
2:33:17 if ever.
2:33:17 And I certainly think they’re not the way that we’re describing them now.
2:33:21 Well, creating existential trauma is one of the things that makes life fun, I guess.
2:33:28 Yeah, it’s just what we do to ourselves.
2:33:30 It gives us really exciting big problems to solve.
2:33:34 Yeah, for sure.
2:33:35 Do you think we will see these AI systems become conscious or convinces that they’re conscious
2:33:42 and then maybe we’ll have relationships with them, romantic relationships?
2:33:47 Well, I think people are going to have romantic relationships with them.
2:33:50 And I also think that some people will be convinced already that they’re conscious.
2:33:55 But I think in order, what does it take to convince people that something is conscious?
2:34:05 I think that we actually have to have an idea of what we’re talking about.
2:34:08 We have to have a theory that explains when things are conscious or not, that’s testable.
2:34:14 And we don’t have one right now.
2:34:16 So I think until we have that, it’s always going to be this sort of gray area where some people
2:34:20 think it hasn’t and some people think it doesn’t.
2:34:22 Because we don’t actually know what we’re talking about that we think it has.
2:34:25 So do you think it’s possible to get out of the gray area
2:34:28 and really have a formal test for consciousness?
2:34:30 For sure.
2:34:31 And for life as you were?
2:34:33 For sure.
2:34:34 As we’ve been talking about for some reason.
2:34:35 Yeah.
2:34:36 Consciousness is a tricky one.
2:34:38 It is a tricky one.
2:34:39 I mean, that’s why it’s called the hard problem of consciousness because it’s hard.
2:34:42 And it might even be outside of the purview of science,
2:34:45 which means that we can’t understand it in a scientific way.
2:34:48 There might be other ways of coming to understand it.
2:34:50 But those may not be the ones that we necessarily want for technological utility
2:34:55 or for developing laws with respect to,
2:34:59 because the laws are the things that are going to govern the technology.
2:35:03 Well, I think that’s actually where the hard problem of consciousness,
2:35:08 a different hard problem of consciousness is that I fear that humans will resist.
2:35:17 That’s the last thing they will resist is calling something else conscious.
2:35:21 Oh, that’s interesting.
2:35:22 I think it depends on the culture though,
2:35:24 because I mean, some cultures already think like everything’s imbued with
2:35:27 you know, a life essence or kind of conscious.
2:35:31 I don’t think those cultures have nuclear weapons.
2:35:34 No, they don’t.
2:35:34 And they’re probably not building the most advanced technologies.
2:35:37 The cultures that are primed for destroying the other,
2:35:41 constructing a very effective propaganda machines of what the other is.
2:35:46 The group to hate are the other cultures that I worry would be very resistant to label something.
2:35:59 To sort of acknowledge the consciousness laid in a thing that was created by us humans.
2:36:06 And so what do you think the risks are there that the conscious things will get
2:36:09 angry with us and fight back?
2:36:11 No, that we would torture and kill conscious beings.
2:36:15 Oh, yeah, I think we do that quite a lot.
2:36:19 Anyway, without, I mean, I don’t, I mean, it goes back to your,
2:36:24 and I don’t know how to feel about this.
2:36:27 But you know, like we talked already about the predator prey thing that like,
2:36:30 in some sense, you know, being alive requires eating other things that are alive.
2:36:36 And even if you’re a vegetarian or, you know, like try to have like your,
2:36:39 like you’re still eating living things.
2:36:41 So maybe part of the story of earth will involve a predator prey dynamic between
2:36:48 humans and human creations.
2:36:52 Yeah, and all of that is part of the time.
2:36:55 But I don’t like thinking about them as like our technologies as a separate species,
2:36:59 because this again goes back to this sort of levels of selection issue.
2:37:02 And you know, if you think about humans individually alive,
2:37:07 you miss the fact that societies are also alive.
2:37:10 And so I think about it much more in the sense of an ecosystem is not the right word,
2:37:18 but we don’t have the right words for these things of like,
2:37:20 and this is why I talk about the technosphere.
2:37:22 It’s a system that is both human and technological.
2:37:26 It’s not human or technological.
2:37:28 And so this is the part that I think we’re really good for the like,
2:37:36 and this is driving in part a lot of the sort of attitude of like,
2:37:40 I’ll kill you first with my nuclear weapons.
2:37:43 We’re really good at identifying things as other.
2:37:47 We’re not really good at understanding when we’re the same or when we’re part
2:37:50 of an integrated system that’s actually functioning together in some kind of cohesive way.
2:37:54 So even if you look at like, you know, the division in American politics or something,
2:37:58 for example, it’s important that there’s multiple sides that are arguing with each other,
2:38:02 because that’s actually how you resolve society’s issues.
2:38:05 It’s not like a bad feature.
2:38:06 I think like some of the sort of extreme positions and like the way people talk about
2:38:10 are maybe not ideal, but that’s how societies solve problems.
2:38:16 What it looks like for an individual is really different than the societal level outcomes
2:38:19 and the fact that like, there is, I don’t want to call it competition or computation.
2:38:24 I don’t know what you call it, but like there is a process playing out in the dynamics of
2:38:28 societies that we are all individual actors in.
2:38:31 And like, we’re not part of that, you know, like it requires all of us acting individually,
2:38:36 but like this higher level structure is playing out some things and like,
2:38:40 things are getting solved for it to be able to maintain itself.
2:38:42 And that’s the level that our technologies live at.
2:38:46 They don’t live at our level.
2:38:47 They live at the societal level and they’re deeply integrated with the social organism,
2:38:52 if you want to call it that.
2:38:54 And so I really get upset when people talk about the species of artificial intelligence.
2:39:00 I’m like, you mean we live in an ecosystem of all these kind of intelligent things and
2:39:04 these animating technologies that were, you know, in some sense, helping to come alive.
2:39:09 We are, we are generating them, but it’s not like the biosphere
2:39:12 eliminated all of its past history when it invented a new species.
2:39:16 All of these things get scaffolded.
2:39:18 And we’re also augmenting ourselves at the same time that we’re building technologies.
2:39:21 I don’t think we can anticipate what that system is going to look like.
2:39:24 So in some fundamental way, you always want to be thinking about the planet as one organism.
2:39:29 The planet is one living thing.
2:39:30 What happens when it becomes multi-planetary?
2:39:33 Is it still just the,
2:39:35 Still the same causal chain.
2:39:37 Same causal chain.
2:39:37 It’s like when the first cell split into two.
2:39:39 That’s what I was talking about when the, when a planet reproduces itself,
2:39:42 the technosphere emerges enough understanding.
2:39:45 It’s, it’s like this recursive, like the entire history of life is just recursion, right?
2:39:50 So you have an original life event.
2:39:52 It evolves for four billion years, at least on our planet.
2:39:55 It evolves a technosphere.
2:39:56 The technologies themselves start to become having this property we call life,
2:40:02 which is the phase we’re undergoing now.
2:40:03 It solves the origin of itself.
2:40:07 And then it figures out how that process all works,
2:40:10 understands how to make more life,
2:40:11 and then can copy itself onto another planet.
2:40:13 So the whole structure can reproduce itself.
2:40:15 And so the original life is happening again right now.
2:40:20 On this planet, in the technosphere,
2:40:22 with the way that our planet is undergoing another transition,
2:40:25 just like at the original life when geochemistry transitioned to biology,
2:40:28 which is the global, for me, it was a planetary scale transition.
2:40:31 It was a multi-scale thing
2:40:33 that happened from the scale of chemistry all the way to planetary cycles.
2:40:36 It’s happening now, all the way from individual humans to the internet,
2:40:41 which is a global technology and all the other things.
2:40:44 Like there’s this multi-scale process that’s happening and transitioning us globally.
2:40:47 And it’s a dramatic transition.
2:40:49 It’s happening really fast.
2:40:51 And we’re living in it.
2:40:53 You think this technosphere that we’ve created,
2:40:55 this increasingly complex technosphere will spread to other planets?
2:40:59 I hope so.
2:41:00 I think so.
2:41:01 You think we’ll become a Type II Kardashev civilization?
2:41:05 I don’t really like the Kardashev scale.
2:41:07 And it goes back to, I don’t like a lot of the narratives about life,
2:41:11 because they’re very survival of the fittest,
2:41:15 energy consuming, this, that, and the other thing.
2:41:17 It’s very, I don’t know, sort of old world, conqueror mentality.
2:41:21 What’s the alternative to that exactly?
2:41:24 I mean, I think it does require life to use new energy sources
2:41:29 in order to expand the way it is.
2:41:31 So that part’s accurate.
2:41:32 But I think this sort of process of life,
2:41:35 like being the mechanism that the universe creatively expresses itself,
2:41:41 generates novelty, explores the space of the possible
2:41:45 is really the thing that’s most deeply intrinsic to life.
2:41:48 And so, you know, these sort of energy consuming scales of technology,
2:41:53 I think is missing the sort of actual feature that’s most prominent
2:41:58 about any alien life that we might find,
2:42:01 which is that it’s literally our universe,
2:42:04 our reality trying to creatively express itself
2:42:07 and trying to find out what can exist and trying to make it exist.
2:42:09 See, but past a certain level of complexity,
2:42:11 unfortunately, maybe you can correct me,
2:42:13 but we’re built, all complex life on Earth
2:42:16 is built on a foundation of that predator/prey dynamic.
2:42:19 Yes.
2:42:20 And so, like, I don’t know if we can escape that.
2:42:22 No, we can’t.
2:42:23 But this is why I’m okay with having a finite lifetime.
2:42:26 And, you know, one of the reasons I’m okay with that, actually,
2:42:29 goes back to this issue of the fact that we’re resource bound.
2:42:33 We live in a, you know, like, we have a finite amount of material,
2:42:37 whatever way you want to define material, I think, like, for me,
2:42:40 you know, material is time, material is information,
2:42:43 but we have a finite amount of material.
2:42:46 If time is a generating mechanism, it’s always going to be finite
2:42:49 because the universe is, you know, like, it’s a resource
2:42:53 that’s getting generated, but it has a size,
2:42:55 which means that all the things that could exist don’t exist.
2:43:01 And in fact, most of them never will.
2:43:03 So, death is a way to make room in the universe
2:43:05 for other things to exist that wouldn’t be able to exist otherwise.
2:43:08 So, if the universe, over its entire temporal history,
2:43:10 wants to maximize the number of things, wants as a hard word,
2:43:13 maximize a hard word, all these things are approximate,
2:43:15 but wants to maximize the number of things that can exist,
2:43:19 the best way to do it is to make recursively embedded stacked objects
2:43:22 like us that have a lot of structure and a small volume of space,
2:43:26 and to have those things turn over rapidly
2:43:28 so you can create as many of them as possible.
2:43:31 So, for sure, there’s a bunch of those kinds of things throughout the universe.
2:43:35 Hopefully.
2:43:36 Hopefully, our universe is teeming with life.
2:43:38 This is, like, early on in the conversation you mentioned
2:43:41 that we really don’t understand much.
2:43:44 Like, there’s mystery all around us.
2:43:47 Yes.
2:43:48 If you had to, like, bet money at it, like, what percent?
2:43:50 So, like, say, a million years from now, the story of science
2:43:58 and human understanding, understanding that started on Earth is written.
2:44:05 Like, what chapter are we on?
2:44:07 Are we, like, is this, like, 1%, 10%, 20%, 50%, 90%?
2:44:13 How much do we understand?
2:44:15 Like, the big stuff.
2:44:16 Not, like, the details of, like, big, important questions and ideas.
2:44:24 I think we’re in our 20s.
2:44:28 And no, like, age-wise, let’s say we’re in our 20s,
2:44:32 but the lifespan is going to keep getting longer.
2:44:34 You can’t do that.
2:44:36 I can.
2:44:37 You know why I use that, though?
2:44:38 I’ll tell you why.
2:44:39 Why my brain went there is because, you know,
2:44:42 anybody that gets an education in physics, you know,
2:44:45 has this sort of trope about how all the great physicists
2:44:48 did their best work in their 20s.
2:44:51 And then you don’t do any good work after that.
2:44:53 And I always thought it was kind of funny,
2:44:55 because for me, physics is not complete.
2:45:02 It’s not nearly complete.
2:45:03 But most physicists think that we understand
2:45:06 most of the structure of reality.
2:45:07 And so I think I actually, I think I put this in the book somewhere,
2:45:12 but, like, this idea to me that societies would discover
2:45:17 everything while they’re young is very consistent
2:45:19 with the way we talk about physics right now.
2:45:21 But I don’t think that’s actually the way that things are going to go.
2:45:25 And you’re finding that people that are making major discoveries
2:45:30 are getting older in some sense than they were.
2:45:32 And our lifespan is also increasing.
2:45:34 So I think there is something about age and your ability to learn
2:45:39 and how much of the world you can see that’s really important
2:45:42 over a human lifespan, but also over the lifespan of societies.
2:45:46 And so I don’t know how big the frontier is.
2:45:49 I don’t actually think it has a limit.
2:45:51 I don’t believe in infinity as a physical thing,
2:45:55 but I think as a receding horizon,
2:45:59 I think because the universe is getting bigger,
2:46:00 you can never know all of it.
2:46:01 Well, I think it’s about 1.7 percent.
2:46:06 1.7, where does that come from?
2:46:09 And it’s a finite, I don’t know, I just made it up.
2:46:11 But it’s like–
2:46:12 That number had to come from somewhere.
2:46:13 Certainly.
2:46:15 I think seven is the thing that people usually pick.
2:46:18 Seven percent?
2:46:18 So I wanted to say 1%, but I thought it would be funnier
2:46:22 to add a point, inject a little humor in there.
2:46:26 So the seven is for the humor.
2:46:28 One is for how much mystery I think there is out there.
2:46:32 99% mystery, 1% known.
2:46:35 In terms of really big, important questions,
2:46:37 say there’s going to be like 200 chapters.
2:46:41 Like the stuff that’s going to remain true.
2:46:43 But you think the book has a finite size.
2:46:47 Yeah, yeah.
2:46:48 And I don’t.
2:46:49 I mean, not that I believe in infinities,
2:46:52 but I don’t, I think the size of the book is growing.
2:46:55 Well, the fact that the size of the book is growing
2:46:59 is one of the chapters in the book.
2:47:01 Oh, there you go.
2:47:03 Oh, we’re being recursive.
2:47:04 I think you have to, you can’t have an ever-growing book.
2:47:09 Yes, you can.
2:47:10 I mean, you just, I mean, I don’t even, because then–
2:47:14 Well, you couldn’t have been asking this
2:47:16 at the original life, right?
2:47:17 Because obviously like, you wouldn’t have existed
2:47:19 at the original life.
2:47:19 But like the question of intelligence
2:47:21 and artificial general, like those questions
2:47:24 did not exist then.
2:47:25 And so, and they in part existed
2:47:29 because the universe invented a space
2:47:30 for those questions to exist through evolution.
2:47:34 But like, I think that question will still stand
2:47:38 a thousand years from now.
2:47:39 It will, but there will be other questions
2:47:41 we can’t anticipate now that we’ll be asking.
2:47:43 Yeah, and maybe we’ll develop the kinds of languages
2:47:47 that we’ll be able to ask much better questions.
2:47:48 Right, or like the theory of like gravitation, for example,
2:47:52 like when we invented that theory,
2:47:53 like we only knew about the planets in our solar system, right?
2:47:56 And now, you know, many centuries later,
2:47:58 we know about all these planets around other stars
2:47:59 and black holes and other things
2:48:01 that we could never have anticipated.
2:48:02 So, and then we can ask questions about them.
2:48:05 You know, like we wouldn’t have been asking
2:48:07 about singularities and like,
2:48:09 can they really be physical things in the universe
2:48:11 several hundred years ago?
2:48:12 That question couldn’t exist.
2:48:15 Yeah, but it’s not.
2:48:17 I still think those are chapters in the book.
2:48:19 Like, I don’t get a sense from that.
2:48:22 So, do you think the universe has an end?
2:48:23 If you think it’s a book with an end?
2:48:26 I think the number of words required
2:48:30 to describe how the universe works has an end, yes.
2:48:33 Meaning, like, I don’t care if it’s infinite or not.
2:48:39 Right.
2:48:39 As long as the explanation is simple and it exists.
2:48:43 Oh, I see.
2:48:44 And I think there is a finite explanation
2:48:47 for each aspect of it.
2:48:48 The consciousness, the life.
2:48:51 Yeah.
2:48:51 I mean, very probably there’s like some…
2:48:55 The black hole thing is like, what’s going on there?
2:49:00 Where’s that going?
2:49:01 It’s so kind of weird.
2:49:01 Like, where do they, what?
2:49:02 And then, you know, why the big bang?
2:49:05 Like, what?
2:49:06 Right.
2:49:07 It’s probably there’s just a huge number of universes
2:49:10 and it’s like, universes inside the universe.
2:49:11 You think so?
2:49:12 I think universes inside universes is maybe possible.
2:49:16 I just think it’s, every time we assume this is all there is,
2:49:23 it turns out there’s much more.
2:49:25 The universe is a huge place.
2:49:27 And we mostly talked about the past and the richness of the past,
2:49:30 but the future, I mean, with many worlds,
2:49:32 interpretation of quantum mechanics.
2:49:34 So…
2:49:36 Oh, I am not a many worlds person.
2:49:37 You’re not.
2:49:38 No, are you?
2:49:39 How many lexes are there?
2:49:41 Depending on the day.
2:49:42 Well…
2:49:43 Do some of them wear yellow jackets?
2:49:44 At the moment, at the moment we asked the question,
2:49:46 there was one at the moment I’m answering it.
2:49:49 There’s now an near infinity, apparently.
2:49:52 I mean, the future is, the future is bigger than the past, yes?
2:49:58 Yes.
2:49:58 Okay.
2:49:59 I think so.
2:49:59 In the past, according to the future, it’s already gigantic.
2:50:02 Yeah.
2:50:03 But yeah, I mean, that’s consistent with many worlds, right?
2:50:05 Because like, there’s this constant branching.
2:50:06 So, but it doesn’t really have a directionality to it.
2:50:10 It’s a, I don’t know, many worlds is weird.
2:50:12 So, my interpretation of reality is like,
2:50:15 if you fold it up, like all that bifurcation of many worlds,
2:50:18 and you just fold it into the structure that is you,
2:50:20 and you just said you are all of those many worlds,
2:50:23 and like, that sort of, you know, like, your history converged on you,
2:50:27 but you’re actually an object exists that’s like, that was selected to exist,
2:50:33 and you’re self-consistent with the other structures.
2:50:36 So, like, the quantum mechanical reality is not the one that you live in.
2:50:39 It’s this very deterministic, classical world.
2:50:43 And you’re carving a path through that space,
2:50:47 but I don’t think that you’re constantly branching into new spaces.
2:50:51 I think you are that space.
2:50:53 Wait, so to you at the bottom, it’s deterministic.
2:50:56 I thought you said the universe.
2:50:57 No, it’s random at the bottom, right?
2:50:59 But like, this randomness that we see at the bottom of reality,
2:51:02 that is quantum mechanics.
2:51:04 I think, like, people have assumed that that is reality.
2:51:07 And what I’m saying is, like, all those things you see in many worlds,
2:51:11 all those versions of you, just collect them up and bundle them up,
2:51:15 and like, they’re all you.
2:51:17 And what has happened is, you know, like, elementary particles don’t have,
2:51:21 they don’t live in a deterministic universe,
2:51:23 the things that we study in quantum experiments.
2:51:26 They live in this fuzzy random space.
2:51:28 But as that structure collapsed and started to build structures
2:51:31 that were deterministic and evolved into you,
2:51:34 you are a very deterministic macroscopic object.
2:51:38 And you can look down on that universe that doesn’t have time in it,
2:51:40 that random structure.
2:51:41 And you can see that all of these possibilities look possible,
2:51:46 but they don’t look, they’re not possible for you,
2:51:48 because you’re constrained by this giant, like, causal structural history.
2:51:52 So you can’t live in all those universes.
2:51:56 You’d have to go all the way back to the very beginning of the universe
2:52:01 and retrace everything again to be a different you.
2:52:03 So where’s the source of the free will for the macro object?
2:52:05 It’s the fact that you’re a deterministic structure living in a random background.
2:52:12 And also all of that selection bundled in you allows you to select on possible futures.
2:52:16 So that’s where your will comes from.
2:52:18 And there’s just always a little bit of randomness,
2:52:20 because the universe is getting bigger.
2:52:22 And, you know, like, this idea that the past is,
2:52:26 and the present is not large enough yet to contain the future,
2:52:29 the extra structure has to come from somewhere.
2:52:33 And some of that is because outside of those giant causal structures
2:52:37 that are things like us, it’s fucking random out there.
2:52:41 And it’s scary.
2:52:44 And we’re all hanging on to each other,
2:52:45 because the only way to hang on to each other,
2:52:47 like, the only way to exist is to, like, cling on
2:52:50 to all of these causal structures that we happen to co-inhabitate existence with
2:52:54 and try to keep reinforcing each other’s existence.
2:52:57 All the selection bundled in.
2:53:00 And not enough in us, but free will is totally consistent with that.
2:53:03 I don’t know what I think about that.
2:53:05 That’s complicated to imagine.
2:53:06 Just that little bit of randomness is enough.
2:53:09 Okay.
2:53:10 Well, it’s also, it’s not just the randomness.
2:53:13 There’s two features.
2:53:13 One is the randomness helps generate some novelty and some flexibility.
2:53:17 But it’s also that, like, because you’re the structure that’s deep in time,
2:53:21 you have this combinatorial history that’s you.
2:53:24 And I think about time and assembly theory, not as linear time, but as combinatorial time.
2:53:31 So if you have all of the structure that you’re built out of,
2:53:34 you, in principle, you know, your future can be combinations of that structure.
2:53:40 You obviously need to persist yourself as a coherent you.
2:53:43 So you want to optimize for a common,
2:53:45 like a future in that combinatorial space that still includes you.
2:53:52 Most of the time for most of us.
2:53:53 And when you make those kinds, and then that gives you a space to operate in.
2:54:01 And that’s your sort of horizon where your free will can operate.
2:54:05 And your free will can’t be instantaneous.
2:54:07 So for, like, example, like I’m sitting here talking to you right now,
2:54:11 I can’t be in the UK and I can’t be in Arizona, but I could plan.
2:54:15 I could execute my free will over time because free will is a temporal feature of life
2:54:21 to be there, you know, tomorrow or the, or the next day if I wanted to.
2:54:24 But what about like the instantaneous decisions you’re making?
2:54:27 Like to, I don’t know, to put your hand on the table.
2:54:31 That’s, I think those were already decided a while ago.
2:54:34 I don’t think they’re the, I don’t think free will is ever instantaneous.
2:54:37 But I know longer time horizon, yep.
2:54:41 There’s some kind of staring going on.
2:54:43 And who’s doing the staring?
2:54:47 You are.
2:54:49 And you being this macro object that’s
2:54:51 encompasses.
2:54:54 Or you being Lex.
2:54:55 Whatever you want to call it.
2:54:58 There, there you are saying words to things once again.
2:55:04 I know.
2:55:05 Why does anything exist at all?
2:55:07 You’ve kind of taken that as a starting point.
2:55:12 Yeah, it exists.
2:55:13 I think that’s the hardest question.
2:55:15 Isn’t it just hard questions stack on top of each other?
2:55:18 Wouldn’t it be the same kind of question of what is life?
2:55:21 It is the same.
2:55:23 I, well, that’s the sort of like, I try to fold all of the questions
2:55:26 into that question.
2:55:26 Cause I think that one’s really hard.
2:55:28 And I think the nature of existence is really hard.
2:55:30 You think actually like answering what is left will help us understand existence.
2:55:34 Maybe, maybe there’s.
2:55:35 It’s turtles all the way down.
2:55:38 And it’ll just understand the nature of turtles will help us kind of march down.
2:55:43 Even if we don’t have the experimental methodology of reaching before the big
2:55:48 bang.
2:55:49 Right.
2:55:50 So, well, I think there’s, there’s sort of two questions embedded here.
2:55:53 I think the one that we can’t answer by answering life is why certain things exist
2:55:58 and others don’t.
2:55:59 But I think the sort of ultimate question, the sort of like prime mover question
2:56:05 of why anything exists, we will not be able to answer.
2:56:08 What’s outside the universe?
2:56:11 Oh, there’s nothing outside the universe.
2:56:13 So I have a very, I am a very, like, I am like the most physicalist that like anyone
2:56:21 could be.
2:56:22 So like for me, everything exists in our universe.
2:56:25 And it, and like, I like to like think it like everything exists here.
2:56:31 So even when we talk about the multiverse, I don’t like, to me, it’s not like there’s
2:56:35 all these other universes outside of our universe that exists.
2:56:38 The multiverse is a concept that exists in human minds here.
2:56:42 And it allows us to have some counterfactual reasoning to reason about our own cosmology.
2:56:48 And therefore, it’s causal in our biosphere to understanding the reality that we live
2:56:52 in and building better theories.
2:56:54 But I don’t think that the multiverse is something like, and also math, like, I don’t
2:56:59 think there’s a platonic world that mathematical things live in.
2:57:02 I think mathematical things are here on this planet.
2:57:05 Like, I don’t think it makes sense to talk about things that exist outside of the universe.
2:57:10 If you’re talking about them, you’re already talking about something that exists inside
2:57:13 the universe and is part of the universe and is part of like what the universe is building.
2:57:17 It all originates here.
2:57:19 It all exists here in some way.
2:57:20 I mean, what else would there be?
2:57:21 That could be things you can’t possibly understand outside of all of this that we call the universe.
2:57:28 And you can say that, and that’s an interesting philosophy.
2:57:30 But again, this is sort of like pushing on the boundaries of like the way that we understand
2:57:34 things.
2:57:35 I think it’s more constructive to say the fact that I can talk about those things is telling
2:57:39 me something about the structure of where I actually live and where I exist.
2:57:42 Just because it’s more constructive doesn’t mean it’s true.
2:57:45 Well, it may not be true.
2:57:49 It may be something that allows me to build better theories I can test to try to understand
2:57:53 something objective.
2:57:54 And in the end, that’s a good way to get to the truth.
2:57:58 Exactly.
2:57:59 Even if you realize you were wrong in the past.
2:58:02 Yeah.
2:58:02 So there’s no such thing as experimental platonism.
2:58:05 But if you think math is an object that emerged in our biosphere, you can start experimenting
2:58:10 with that idea.
2:58:11 And that, to me, is really interesting.
2:58:14 Like to think about, well, I mean, mathematicians do think about math.
2:58:18 Sometimes it’s experimental science.
2:58:19 But to think about math itself as an object for study by physicists rather than a tool
2:58:28 physicists use to describe reality, it becomes the part of reality they’re trying to describe
2:58:33 to me as a deeply interesting inversion.
2:58:35 What to use most beautiful about this kind of exploration of the physics of life that
2:58:41 you’ve been doing?
2:58:42 I love the way it makes me feel.
2:58:45 And then you have to try to convert the feelings into visuals and the visuals into words?
2:58:53 Yeah.
2:58:54 So I think I love the way it makes me feel to have ideas that I think are novel.
2:59:03 And I think that the dual side of that is the painful process of trying to communicate that
2:59:07 with other human beings to test if they have any kind of reality to them.
2:59:11 And I also love that process.
2:59:15 I love trying to figure out how to explain really deep abstract things that I don’t think
2:59:21 that we understand and trying to understand them with other people.
2:59:24 And I also love the shock value of this kind of idea we were talking about before of being
2:59:32 on the boundary of what we understand.
2:59:34 And so people can kind of see what you’re seeing, but they haven’t ever sought that way before.
2:59:39 And I love the shock value that people have, that immediate moment of recognizing that there’s
2:59:45 something beyond the way that they thought about things before.
2:59:48 And being able to deliver that to people I think is one of the biggest joys that I have.
2:59:52 And maybe it’s that sense of mystery, to share that there’s something beyond the frontier of
2:59:58 how we understand and we might be able to see it.
3:00:00 And you get to see the humans transformed by the new idea?
3:00:04 Yes. And I think my greatest wish in life is to somehow contribute to an idea that transforms
3:00:13 the way that we think. I have my problem I want to solve, but the thing that gives
3:00:19 me joy about it is really changing something and ideally getting to a deeper understanding of
3:00:27 how the world works and what we are.
3:00:29 Yeah, I would say understanding life at a deep level is probably one of the most exciting
3:00:38 problems, one of the most exciting questions.
3:00:40 So I’m glad you’re trying to answer just that and doing it in style.
3:00:47 It’s the only way to do anything.
3:00:49 Thank you so much for this amazing conversation. Thank you for being you, Sarah.
3:00:54 This was awesome.
3:00:56 Thanks, Lex.
3:00:57 Thanks for listening to this conversation with Sarah Walker.
3:01:00 To support this podcast, please check out our sponsors in the description.
3:01:04 And now let me leave you with some words from Charles Darwin.
3:01:07 In the long history of humankind and animal kind too, those who learned to collaborate
3:01:15 and improvise most effectively have prevailed.
3:01:18 Thank you for listening and hope to see you next time.
3:01:26 [Music]