Sara Walker: The Origin of Life on Earth and Alien Worlds
Guests: Sara Walker
Full transcript · Read the full conversation in order, with its original context.
Contents15 chapters
§01Transcript · 00:00–08:23§02Transcript · 08:28–16:23§03Transcript · 16:26–24:33§04Transcript · 24:36–32:48§05Transcript · 32:49–40:32§06Transcript · 40:34–49:02§07Transcript · 49:04–56:48§08Transcript · 56:49–1:04:21§09Transcript · 1:04:25–1:12:36§10Transcript · 1:12:38–1:20:18§11Transcript · 1:20:20–1:28:23§12Transcript · 1:28:24–1:35:55§13Transcript · 1:36:02–1:44:48§14Transcript · 1:44:53–1:53:30§15Transcript · 1:53:32–1:59:21§01Transcript · 00:00–08:23
The following is a conversation with Sarah Walker,
an astrobiologist and theoretical physicist
at Arizona State University and the Santa Fe Institute.
She's interested in the origin of life,
how to find life on other worlds,
and in general, the more fundamental question
of what even life is.
She seeks to discover the universal laws
that describe living systems on earth and elsewhere
using physics, biology, and computation.
Quick mention of our sponsors, Athletic Greens,
NetSuite, Blinkist, and Magic Spoon.
Check them out in the description to support this podcast.
As a side note, let me say that my hope for this podcast
is to try and alternate between technical
and non-technical discussions,
to jump from the big picture
down to specific detailed research
and back to the big picture,
and to do so with scientists and non-scientists.
Long-term, I hope to alternate between discussions
of cutting-edge research in AI, physics, biology,
to topics of music, sport, and history,
and then back to AI.
AI is home.
I hope you come along with me
for that wild, oscillating journey.
Some people message me saying to slow down
since they're falling behind on the episodes
of this podcast.
To their disappointment, I have to say
that I'll probably do more episodes, not less,
but you really don't need to listen to every episode.
Just listen to the ones that spark your curiosity.
Think about it like a party full of strangers.
You don't have to talk to everyone.
Just walk over to the ones who look interesting
and get to know them.
And if you're lucky, that one conversation with a stranger
might change the direction of your life.
And it's a short life, so be picky with the strangers
you talk to at this metaphorical party.
This is the Lux Friedman Podcast,
and here is my conversation with Sarah Walker.
How did life originate on Earth?
What are the various hypotheses
for how life originated on Earth?
- Yeah, so I guess you're asking a historical question,
which is always a good place to start thinking about life.
So there's a lot of ideas about how life started on Earth.
Probably the most popular is what's called
the RNA world scenario.
So this idea is probably the one
that you'll see most reported in the news,
and is based on the idea that there are
molecules in our bodies that relay genetic information,
and we know those as DNA, obviously,
but there's also sort of an intermediary
called RNA, ribonucleic acid,
that also plays the role of proteins.
And people came up with this idea in the '80s
that maybe that was the first genetic material
because it could play both roles
of being genetic and performing catalysis.
And then somehow that idea got reduced
to this idea that there was a molecule
that emerged on early Earth
and underwent Darwinian evolution,
and that was the start of life.
So there's a lot of assumptions packed in there
that we could unpack,
but that's sort of the leading hypothesis.
There's also other ideas about life starting as metabolism,
and so that's more connected
to the geochemistry of early Earth,
and would be kind of more focused on this idea
that you get some kind of catalytic cycle
of molecules that can reproduce themselves
and form some kind of metabolism,
and then life starts basically as self-organization,
and then you have to explain how evolution comes later.
- Right, so that's the difference
between sort of energy and genetic code?
So like energy and information,
are those the two kind of things there?
- Yeah, I think that's a good way of putting it.
It's kind of funny 'cause I think most of the people
that think about these things are really disciplinary biased,
so the people that tend to think about genetics
come from a biology background
and they're really evolution-focused,
and so they're worried about
where does the information come from
and how does it change over time,
but they're talking about information in a really narrow way
where they're talking about a genetic sequence,
and then most of the people that think about metabolism,
origins of life scenarios,
tend to be people like physicists or geochemists
that are worried about what are the energy sources
and what kinds of organization
can you get out of those energy sources.
- Okay, so which one's your favorite?
- I don't like either.
- Okay, can we talk about them
for a little bit longer though?
- Yeah, no, that's fine.
(laughing)
- So okay, so there's early Earth.
What was that like?
Was there just mostly covered by oceans?
Was there heat sources, energy sources?
So if we talk about the metabolism view
of the origin of life, where was the source of energy?
- Probably the most popular view
for where the origin of life happened on Earth
is hydrothermal vents because they had sufficient energy,
and so we don't really know a lot about early Earth.
We have some ideas about when oceans first formed
and things like that, but the time of the origin of life
is kind of not well understood or pinned down,
and the conditions on Earth at that time are not well known.
But a lot of people do think
that there was probably hydrothermal vents,
which are really hot, chemically active regions,
say on the sea floor in modern times,
which also would have been present on early Earth,
and they would have provided energy and organics
and basically all of the right conditions
for the origins of life, which is one of the reasons
that we look for these hydrothermal systems
when we're talking about life elsewhere too.
- Okay, and for the genetic code,
the idea is that the RNA is the first,
like why would RNA be the first moment you can say it's life?
I guess the idea is it could both
have persistent information,
and then it can also do some of the work of like, what,
creating a self-sustaining organism?
- Yeah, that's the basic idea.
So the idea is you have, in an RNA molecule,
you have a sequence of characters, say,
so you can treat it like a string in a computer,
and it can be copied, so information can be propagated,
which is important for evolution,
because evolution happens
by having inheritance of information.
So for example, like my eyes are brown
because my mother's eyes were brown.
So you need that copying of information,
but then you also have the ability to perform catalysis,
which means that that RNA molecule
is not inert in that environment,
but it actually interacts with something
and could potentially mediate, say, a metabolism
that could then fuel the actual reproduction
of that molecule.
So in some ways, people think that RNA gives you,
you know, the most bang for your buck in a single molecule,
and therefore, you know, it gives you all the features
that you might think are life.
And so this is sort of where this RNA world conjecture
came from, is because of those two properties.
- Isn't it amazing that RNA came to be in general?
Isn't it? - Yes, that is amazing.
- Okay, so we're not talking down about RNA.
- No, no, I love RNA.
It's one of my favorite molecules.
I think it's beautiful. - It's just not step one.
- Yeah, I think the issue,
it's not even the RNA world is a problem.
And actually, if you really dig into it,
the RNA world is not one hypothesis.
It is a set of hypotheses.
And they range from a molecule of RNA spontaneously emerged
on the early Earth and started evolving,
which is kind of like the hardest RNA world scenario,
which is the one I cited,
and I get a little animated about
'cause it seems so blatantly wrong to me,
but that's a separate story.
And then the other one is actually something I agree with,
which is that you can say there was an RNA world
because RNA was the first genetic material
for life on Earth.
So an RNA world could just be the earliest organisms
that had genetics in a modern sense,
didn't have DNA evolved yet, they had RNA.
Right, and so that's sort of a softer RNA world scenario
in the sense that it doesn't mean
it was the first thing that happened,
but it was a thing that definitely was part
of the lineage of events that led to us.
- So if life was like a best of album,
it would be one of the songs on there.
- Yes. - One of the early songs.
Okay.
- It's on the greatest hits.
- Greatest hits, that's the word I was looking for.
Okay, did life, do you think, originate once, twice,
§02Transcript · 08:28–16:23
three times on Earth, multiple times?
What do you think?
- I think that's a really difficult question.
- Is it an important question?
- It's a super important question, no.
No, it's a really important question.
There's a lot of questions in that question.
So one of the first ones that I think needs to be addressed
is is the original life a continuous process on our planet?
So we think about the original life as something
that happened on Earth, say, almost four billion years ago,
because we have evidence of life emerging
very early on our planet.
And then an original life event, quote unquote,
a singular event, whatever that was, happened.
And then all life on Earth that we know
is a descendant of that particular event
in our universe, right?
And so, but we don't have any idea one way or the other
if the original life is happening repeatedly,
and maybe it's just not taking off
because life is already established.
That's an argument that people will make.
Or maybe there are alternative forms of life on Earth
that we don't even recognize.
So this is the idea of a shadow biosphere,
that there actually might just be
completely other life on Earth,
but it's so alien that we don't even know what it is.
- I'm gonna have to talk to you about the shadow biosphere.
- Yeah, that's a fun one.
- In a second, but first, let me ask for the other
alternative, which is panspermia.
- Right.
- So that's the idea, the hypothesis that life exists
elsewhere in the universe and got to us
in like an asteroid or a planetoid or some,
according to Wikipedia, space dust,
whatever the heck that is.
It sounds fun, but basically it rode along
whatever kind of rock and got to us.
Do you think that's at all a possibility?
- Sure, so I think the reason that most
original life scientists are interested
in the original life on Earth and say not the original life
on Mars and then panspermia, the exchange of life
between planets being the explanation,
is once you start removing the original life from Earth,
you know even less about it than you do
if you study it on Earth.
Although I think there are ways of reformulating the problem.
This is why I said earlier, oh, you mean the historical
original life problem, you don't mean the problem
of how does life arise in the universe
and what the universal principles are,
because there's this historic problem,
how did it happen on early Earth?
And there's a more tractable general problem
of how does it happen?
And how does it happen is something
we can actually ask in the lab.
How did it happen on early Earth
is a much more detailed and nuanced question.
It requires detailed knowledge of what was happening
on early Earth that we don't have.
And I'm personally more interested in general mechanisms,
so to me it doesn't matter if it happened on Earth
or it happened on Mars.
It just matters that it happened.
We have evidence it happened.
The question is, did it happen more than once
in our universe?
And so the reason I don't find panspermia
as a particularly, I think it's a fascinating hypothesis.
I definitely think it's possible.
And I in particular think it's possible
once you get to the stage of life where you have technology
because then you obviously can spread out into the cosmos.
But it's also possible for microbes
because we know that certain microorganisms
can survive the journey in space
and they can live in a rock and go between Mars and Earth.
Like people have done experiments
to try to prove that could work.
So in that scenario, it's super cool
'cause then you get planetary exchange.
But say we go look for life on Mars
and it ends up being exactly the same life we have on Earth,
biochemically speaking,
then we haven't really discovered something new
about the universe.
What kind of aliens are possible?
Were there other original life events?
If all the life we ever find
is the same original life event in the universe,
it doesn't help me solve my problem.
- But it's possible that that would be a sign
that you could separate the environment
from the basic ingredients.
- Yes, that's true.
- You can have like a life gun
that you shoot throughout the universe.
And then like once you shoot it,
it's like the Simpsons with a make-up gun.
That was a great episode.
When you shoot this life gun, it'll find the Earth's.
It'll like get sticky.
It'll stick to the Earth's.
And that kind of reduces the barrier
of like the time it takes,
the luck it takes to actually from nothing,
from the basic chemistry,
from the basic physics of the universe
for the life to spring up.
- Yeah, I think this is actually super important
to just think about like does life
getting seated on a planet
have to be geochemically compatible with that planet?
So you're suggesting like we could just shoot guns in space
and like life could go to Mars
and then it would just live there and be happy there.
But that's actually an open question.
So one of the things I was gonna say
in response to your question
about whether the origin of life happened once
or multiple times is for me personally right now
in my thinking, although this changes on a weekly basis,
but is that I think of life more as a planetary phenomena.
So I think the origin of life
because life is so intimately tied to planetary cycles
and planetary processes.
And this goes all the way back
through the history of our planet
that the origin of life itself grew out of geochemistry
and became coupled and controlled geochemistry.
And when we start to talk about life existing on the planet
is when we have evidence of life
actually influencing properties of the planet.
And so if life is a planetary property,
then going to Mars is not a trivial thing
because you basically have to make Mars more Earth-like.
And so in some sense,
like when I think about sort of long-term vision
of humans in space, for example,
really what you're talking about when you're saying
let's send our civilization to Mars
is you're not saying let's send our civilization to Mars,
you're saying let's reproduce our planet on Mars.
Like the information from our planet
actually has to go to Mars and make Mars more Earth-like,
which means that you're now having a reproduction process,
like a cell reproduces itself
to propagate information in the future.
Planets have to figure out how to reproduce their conditions
including geochemical conditions on other planets
in order to actually reproduce life in the universe,
which is kind of a little bit radical.
But I think for long-term sustainability
of life on a planet, that's absolutely essential.
- Okay, so if we were to think about
life as a planetary phenomena,
and so life on Mars would be best
if it's way different than life on Earth,
we have to ask the very basic question of what is life?
- I actually don't think that's the right question to ask.
It took me a long time to get there, right?
- You cross it out?
- Yeah, you cross it off your list, it's wrong.
- Next question.
- No, no, no, I mean, I think it has an answer,
but I think part of the problem is,
most of the places in science where we get really stuck
is because we don't know what questions to ask.
And so you can't answer a question
if you're asking the wrong question.
And I think the way I think about it
is obviously I'm interested in what life is,
so I'm being a little cheeky when I say
that's the wrong question to ask.
That's exactly the question
that's the core of my existence.
But I think the way of framing that is
what is it about our universe that allows features
that we associate life to be there?
And so really what I guess when I'm asking that question,
what I'm after is an explanatory framework
for what life is, right?
And so most people, they try to go in and define life
and they say, well, life is, say,
a self-reproducing chemical system
capable of Darwinian evolution.
That's a very popular definition for life.
Or life is something that metabolizes and eats.
That is not how I think about life.
What I think about life is there are principles
and laws that govern our universe
that we don't understand yet
that have something to do with how information interacts
with the physical world.
I don't know exactly what I mean even when I say that
§03Transcript · 16:26–24:33
because we don't know these rules.
But it's a little bit like, I like to use analogies.
You'll give me time to be like a little long-winded
for a second, even in essay.
But sort of like if you look at the history of physics,
for example, this is like,
so we are in the period of the development
of thought on our planet
where we don't understand what we are yet, right?
There was a period of thought in the history of our planet
where we didn't understand what gravity was.
And we didn't understand, for example,
that the planets in the heavens were actually planets
or that they operated by the same laws that we did.
And so there has been this sort of progression
of getting a deeper understanding
of explaining basic phenomena like,
I'm not gonna drop the cup, I'll drop the water bottle.
There you go.
Okay, that fell, right?
But why did that fall?
This is why I'm a theorist, not an experimentalist.
- That could have gone wrong in so many ways.
- I know, it could have,
especially if I did the cup and it smashed.
Anyway.
So if you think, you take this view
that there's sort of some missing principles,
I associate them to information.
And what the sort of feeling there is,
there's some missing explanatory framework
for how our universe works.
And if we understood that physics,
it would explain what we are.
It might also explain a lot of other features
we don't associate to life.
And so it's a little like people accept the fact
that gravity is a universal phenomena.
But when we wanna study gravity,
we study things like large scale,
galactic structures or black holes or planets.
If we wanna understand information
and how it operates in the physical world,
we study intelligent systems or living systems
because they are the manifestation of that physics.
And the fact that we can't see that clearly yet,
or we don't have that explanatory framework,
I think it's just because we haven't been thinking
about the problem deeply enough.
But I feel like if you're explaining something,
you're deriving it from some more fundamental property.
And of course, I have to say I'm wearing my physicist hat.
So I have a huge bias of liking simple,
elegant explanations of the universe
that really are compelling.
But I think one of the things that I've sort of,
maybe in some ways rejected my training as a physicist
is that most of the elegant explanations
that we have so far don't include us in the universe.
And I can't help but think there's something really special
about what we are,
and there have to be some deep principles at play there.
And so that's sort of my perspective on it.
Now, when you ask me what life is,
I have some ideas of what I think it is.
But I think that we haven't gotten there yet
because we haven't been able to see that structure.
And just to go back to the gravity example,
it's a little like, in ancient times, they didn't know,
I was talking about stars and heavens and things,
they didn't know those were governed
by the same principles as that starting to experiment.
Here's where I was going with it.
Once you realize, like Newton did,
that heavenly motions and earthly motions
are governed by the same principles
and you unify terrestrial and celestial motion,
you get these more powerful ideas.
And I think where life is,
is somehow unifying these abstract ideas
of computation and information
with the physical world, with matter,
and realizing that there's some explanatory framework
that's not physics and it's not computation,
but it's something that's deeper.
- So answering the question of what is life
requires deeply understanding something about the universe
as information processing, the universe as computation.
- Sort of. - It's something about,
like, would, once you come up with an answer
to what is life, will the words information
and computation be in the paragraph?
- No, I don't think so. - God damn it, okay.
- I know, it doesn't help, does it?
I know, I hate, actually, I hate this about what I do
because it's so hard to communicate, right, with words.
Like, when you have words that are ideas
that have historically described one thing
and you're trying to describe something
people haven't seen yet, and the words just don't fit.
- So what's wrong, is it too ambiguous?
The word information?
We could switch to binary if you want.
- Yeah, no, I don't think it's binary either.
I think information's just loaded.
I use it, so the other way I might talk about it
is the physics of causation, but I think that's worse
because causation is an even more loaded word
than information.
- So causation is fundamental, you think?
- I do, yeah, and in some sense, I think the physics,
so this is the really radical part.
Some sense, like, when I really think about it
sort of most deeply, what I think life is
is actually the physics of existence.
What gets to exist and why?
And for simple elementary particles,
that's not very complicated
'cause the interactions are simple,
but for things like you and me and human civilizations,
what comes next in the universe
is really dependent on what came before,
and there's a huge space of possibilities
of things that can exist, and when I say information
and causation, what I mean is,
why is it that cups evolved in the universe
and not some other object that could deliver water
and not spill it?
I don't know what you would call it.
Maybe it wouldn't be a cup, but it's a huge,
people talk about the space of things that could exist
as being actually infinitely large, right?
I don't know if I believe in infinity,
but I do think that there is something very interesting
about the problem of what exists
in its relationship to life.
- So do you think the set of things
that could exist is finite?
So it's very large, but if we were to think
about the physics of existence,
like how many shapes of mugs can there be?
Like is, in the initial programming--
- I should go to the math department for that.
- So that's not a topology question.
I just mean, maybe another way to ask is,
what do you think is fundamental to the universe
and what is emergent?
So if existence, are we supposed to think of that
as somehow fundamental, you think?
- So there's a couple of problems in physics
that I think this is related to.
One is, why does mathematics work
at describing reality so well?
And then there is this problem of,
we don't understand why the laws of physics
are the way they are, or why certain things get to exist,
or what put in place the initial condition of our universe.
There's all of these sort of really deep and big problems,
and they all indirectly are related, I think,
to the same kind of thing that,
our physics is really good
if you specify the initial condition
at specifying a certain sequence of events,
but it doesn't deal with the fact
that other things could have happened,
which is kind of an informational property,
like a counterfactual property.
And it's not good at explaining
this conversation right now.
There are certain things that are outside
the explanatory reach of current physics,
and I think they require looking at it
from a completely different direction.
And so I don't wanna have to fine tune
the initial condition of the universe
to specify precisely all the information
in this conversation.
I think that's a ridiculous assertion.
But that's sort of how people wanna frame it
when they're talking about,
the standard model is sufficient
if we had computing power to basically explain
all of life and our existence.
- An interesting thing you said is,
the way we think about information computation
is by observing a particular kind of systems on Earth
that exhibits something we think of as intelligence.
But that's like looking at, I guess,
the tip of an iceberg,
and we should be really looking at the fundamentals
of the iceberg, like what makes water and ice
and the chemistry from which intelligence emerges.
- Yes, yes. - Essentially.
- We can't just couple the information from the physics,
and I think that's what we've gotten really good at doing,
especially with sort of the modern age
where software is so abstracted from hardware.
But the entire process of biological evolution
has basically been built,
like been building layers of increasing abstraction.
§04Transcript · 24:36–32:48
And so it's really hard to see that physics in us,
but it's much clearer to see it in molecules.
- Yeah, but I guess I'm trying to figure out,
what do you think are the best tools to look at it?
What do you think?
- An open mind?
Is that a tool? (laughs)
- What's the physics of an open mind?
- I think if we solve that, we'll solve everything.
I'm saying an open mind
because I think the biggest stumbling block
to understanding sort of the things
I've been trying to articulate,
or, and when I talk also with colleagues
that are thinking deeply about these same issues,
is none of it is inconsistent with what we know.
It's just such a radically different perception
of the way we understand things now
that it's hard for people to get there.
And in some ways you have to almost forget
what you've learned in order to learn something new, right?
So I feel like most of my career
trying to understand the problem of life
has been variously forgetting and then relearning
things that I learned in physics.
And I think you have to have a capacity to learn things,
but then accept that things that you learned
might not be true.
Or might need refinement or reframing.
And the best way I can say that
is just like with a physics education,
there are just certain things you're told in undergrad
that are like facts about the world.
And your physics professors never tell you
that those facts actually emerge from a human mind, right?
So we're taught to think about,
say the laws of physics, for example,
as this like autonomous thing
that exists outside of our universe
and tells our universe how it works.
But the laws of physics were invented by human minds
to describe things that are regularities
in our everyday experience.
They don't exist autonomous to the universe.
- Right, so it's like turtles on top of turtles,
but eventually it gets to the human mind
and then you have to explain the human mind
with the turtles.
- Yes.
- So you have to, it comes from humans,
this understanding, this simplification of the universe,
these models.
There's a guy named Stephen Wolfram,
there's a concept called cellular automata.
So there's some mysteries in these systems
that are computational in nature
that have maybe echoes of the kind of mysteries
we should need to solve to understand what is life.
So if we could talk, take a computational view of things,
do you think there's something compelling
to reducing everything down to computation,
like the universe is computation,
and then trying to understand life.
So throw away the biology, throw away the chemistry,
throw away even the physics that you learn
undergrad and graduate school,
and more look at these simple little systems,
whether it's cellular automata
or whatever the heck kind of computational systems
that operate on simple local rules
and then create complexity as they evolve.
Is it at all, do you think, productive
to focus on those kinds of systems
to get an inkling of what is life?
And if it is, do you think it's possible
to come up with some kind of laws and principles
about what makes life in those computational systems?
- So I like cellular automata,
I think they're good toy models,
but mostly where I've thought about them and used them
is to actually, let's say, poke at
sort of the current conceptual framework that we have
and see where the flaws are.
So I think the part that you're talking about
that people find intriguing is that
if you have a fairly simple rule
and you specify some initial condition
and you run that rule on that initial condition,
you could get really complex patterns emerging.
And ooh, doesn't that look lifelike?
- Well, it's like really surprising,
isn't it really surprising? - It is really surprising.
And they're beautiful.
And I think they have a lot of nice features
associated to them.
I think the things that I find,
yeah, so I do think as a proof of principle
that you can get complex things
emerging from simple rules, they're great.
As a sort of proof of principle
about some of the ways that we might think of computation
as being sort of a fundamental principle
for dynamical systems and maybe the evolution
of the universe as a whole,
they're a great model system.
As an explanatory framework for life,
I think they're a bit problematic
for the same reason that the laws of physics
are a bit problematic.
And the clearest way I can articulate that
is like cellular automata are actually cast
in sort of a conceptual framework
for how the universe should be described
that goes all the way back to Newton, in fact,
with this idea that we can have a fixed law of motion,
which exists sort of, it's given to you.
The great programmer in the sky gave you this equation
or this rule, and then you just run with it.
And the rule doesn't have,
so a good feature of the rule
is it doesn't have specified in the rule
information about the patterns it generates.
So you wouldn't want, for example,
my cup or my water bottle or me sitting here
to be specified in the laws of physics.
That would be ridiculous 'cause it wouldn't be
a very simple explanation of all the things happening.
It'd have to explain everything.
So, and cellular automata have that feature,
and the laws of physics have that feature.
But you also need to specify the initial condition.
And it also, it basically means that everything that happens
is sort of a consequence of that initial condition.
And I think this kind of framework
is just not the right one for biology.
And part of the way that it's easiest to see this
is a lot of people talk about self-reference
being important in life.
The fact that, you know, like the genome
has information encoded in it.
That information gets read out.
It specifies something about the architecture of a cell.
The architecture of the cell includes the genome.
So the genome has basically self-referential information.
Self-reference obviously comes up in computational law
because it's kind of foundational to Turing's work
and what Gödel did with the incompleteness theorems
and things.
So there's a lot of parallels there,
and people have talked about that at depth.
But the other way of kind of thinking about it
in terms of like a more physics-y way of talking about it
is that what it looks like in biology
is that the rules or the laws depend on the state.
This is typical in computer science.
This is obvious to you.
You know, the update rule depends
on the state of the machine, right?
But you know, you don't think about, you know,
that being sort of the dynamic in physics.
It's, you know, the rules given to you,
and then it, you know, it's a very special subclass,
say, of computations if, you know,
you don't ever change the update.
But in biology, it seems to be that the state
and the law change together as a function of time,
and we don't have that as a paradigm in physics.
And so a lot of people have talked about this
as being kind of a perplexing feature
that maybe there are certain scenarios
where the laws of physics or the laws
that govern a particular system
actually change as a function of state of that system.
- That's trippy.
- Yeah.
- So yeah, the hope of physics,
it's a hope, I guess, but often stated
as a underlying assumption is that the law is static.
- Right.
- Okay.
- And even having laws that vary in time,
not even as a function of the state,
is very radical when you--
- The time in general.
- Yeah.
- Like, you wanna remove time from the equation
as much as possible.
- Yeah, I do.
There's some interesting things in this,
like when we think more deeply about the actual physics
that we're trying to propose governs life,
me with collaborators, and then also other people
that think about similar things,
that time might actually be fundamental,
and there really is an ordering to time.
And that events in the universe are unique
because they have a particular, they happen,
like an object in the universe requires
a certain history of events in order to exist,
which therefore suggests that time
§05Transcript · 32:49–40:32
really does have an ordering.
I'm not talking about the flow of time
and our perception of time,
just the ordering of events.
- Causation of things.
- Yes, causation, there's that word again.
- So causation, that's when you say time,
you mean causation.
- Yes.
- In your proposed model of the physics of life,
the fundamental thing would be causation.
If you were to bet your money on one particular horse
or whatever.
- Yes.
- And then space is emergent?
- Yes.
- So everything's emergent except time.
- Kind of, yeah, or causation.
- And laws change all the time.
Why does it look like laws are the same?
- Well, because, well one way,
and I actually, this idea comes from Lee Cronin
'cause I work with him very closely on these things,
is that the laws of physics look the way they do
because they're low memory laws.
So they don't require a lot of information to specify them.
They're very easy for the universe to implement.
But if you get something like me, for example,
I require a four billion year history
to exist in the universe.
I come with a lot of historical baggage.
And that's part of what I am as a set of causes
that exist in the universe.
So I have local rules that apply to me
that are associated with sort of the information
in my history that aren't universal
to every object in the universe.
And there are some things that are very easy
to implement, low memory rules,
that apply to everything in the universe.
- So there's no shortcuts to you.
- No.
So yeah, I don't believe in things like Boltzmann brains
or fluctuations out of the vacuum
that can produce things like your desk ornaments.
I actually think they require a particular causal chain
of events to exist.
- Well, I appreciate the togetherness of that.
But so how does that, if we have to simulate
the entire universe to create the ornaments
in the two of us, how are we supposed
to create engineer life in a lab?
- Yeah, that's, this goes back to sort of the critique
of the RNA world.
I think one of the problems,
and I'll get to answering your question,
but I think this is kind of relevant here.
One of the problems of the RNA world,
when we test it in the laboratory,
is how much information we're putting into the experiment.
We specify the flask, we make pure reagents,
we mix them, we take them out,
we put them in the next flask, we change the pH,
we change the UV light, and then we get a molecule.
And it's not even an RNA molecule necessarily,
it might just be a base, right?
And so people don't usually think about the fact
that we're agents in the universe making that experiment,
and therefore we put a little bit of life
into that experiment.
Because it's part of our biological lineage,
in the same sense that a cup,
or I am a part of the biological lineage.
The experiment is-- - Our ideas are injecting life.
- Yes. - To the experiment.
- And the constraints that we put on the experiments.
Because those conditions wouldn't exist in the universe,
on planet Earth at that time,
without us as the boundary condition, right?
So-- - Even though we're not actually
adding any actual chemistry or biology
that could be identified as life,
are the constraints we're adding to the experiment,
the design of the experiment.
- Yeah, you can think of the design of the experiment
as a program, you put information in.
It's an algorithmic procedure
that you design the experiment.
And so the origin of life problem becomes one
of minimizing the information we put into physics,
to actually watch the spontaneous origin of life.
- Can we have, so can, is it possible in the lab
to have an information vacuum then?
So like-- - If we could, we would,
that would be amazing, I don't know.
That's a good question for, more for Lee.
- Yeah, you guys, by the way,
for people who don't know Lee Cronin,
is, you guys are colleagues. - Yeah.
- And I've gotten the chance to listen
to the two of you talking.
There's great chemistry,
and you're brilliant at brainstorming together.
And there's a really exciting community here
of brilliant people from different disciplines
working on the problem. - Yeah.
- Of life, of complexity, of, I don't know, whatever.
The words fail us to describe the exact problem
we're trying to actually-- - Right.
- Understand here, intelligence, all those kinds of things.
Okay, so what, from a lab perspective,
so Lee, I guess, would you call him a chemist?
No, what is he? - I think by training,
he's a chemist, but I think most of the people
that work in the field we do have lost their discipline.
(laughing)
That's why I couldn't answer your question earlier.
- Okay. - I don't know what you call him.
- Yeah. - I don't know
what I call myself.
I don't know what I call any of my friends.
- So why is it so hard to create,
and it's an interesting question,
to create biological life in the lab?
Like, from your perspective,
is that an important problem to work on,
to try to recreate the historical origin of life on Earth,
or echoes of the historical origin?
- I think echoes is more appropriate.
I don't think asking the question
of what was the exact historical sequence of events,
and engineering every step in the process
to make exactly the chemistry of life on Earth
as we know it is a meaningful way of asking the question.
And it's a little bit like,
you know, since you're in computer science,
like if you know the answer to a problem,
it's easier to find a program to specify the output, right?
But if you don't know the answer a priori,
you know, finding an algorithm for,
like say finding a prime or something,
it's easy to, you know, verify it's a prime number.
It's hard to find the next prime.
And the way the original life is structured right now,
and the historical problem is,
you know the answer, and you're trying to retrodict it
by breaking it down into the set of procedures
where you're putting a lot of information in.
And what we need to do is ask the question of,
how is it that the rules of how our universe is structured
permit things like life to exist,
and what is the phenomena of life?
And those questions are obviously
essentially the same question.
And so you're looking essentially for this missing physics,
this missing explanation for what we are,
and you need to set up proper experiments
that are gonna allow you to probe
the vast complexity of chemistry in an unconstrained way
with as little information put in as possible
to see when things, when does information actually emerge?
How does it emerge?
What is it?
And part of the sort of conjecture we have is
that this physics only becomes relevant,
or at least this is my personal conjecture,
and it's sort of validated by this kind of theory
experiment collaboration that we have working in this area.
That this, you know, sort of,
I made the point about like gravity existing everywhere,
right, but when you study an atomic nucleus,
you don't care about gravity.
It's not relevant physics there, right?
It's weak, it doesn't matter.
And so this idea that there's kind of a physics
associated with information,
for me, it's very evident that that physics
doesn't become relevant until you need information
to specify the existence of a particular object.
And the scale of reality where that happens is in chemistry
because of the combinatorial diversity of chemical objects
that can exist far out exceeds
the amount of resources in our universe.
So if you want it, you can't make every possible protein
of length, you know, 200 amino acids,
there's not enough resources.
So in order for this particular protein to exist,
and this protein to exist in high abundance,
means that you have to have a system that has knowledge
of the existence of that protein and can build it.
- So existence comes to be at the chemical level.
So existence is most,
is best understood at the chemical level.
- It's most evident.
It's a little bit like,
nobody argues that gravity doesn't exist
in an atomic nucleus, it's just not relevant physics there.
- So the physics of information--
- Is everywhere, it exists at every combinatorial scale,
but it becomes more and more relevant
§06Transcript · 40:34–49:02
the more set of possibilities that could exist.
Because you have to specify more and more
about why this thing exists and not the infinite,
it's not an infinite set, but you know,
the set of, undefined set of other things that could exist.
- So can I ask a weird question?
Which is, so let's look into the future.
- I try that every day, it never works.
- So say a Nobel Prize is given in physics,
maybe chemistry, for discovering the origin of life.
No, but not the historical origin.
Some kind of thing that we're talking about.
What exactly would,
what do you think that,
what do you think that person,
maybe you, did to get that Nobel Prize?
Like what would they have to have done?
'Cause you could do a bunch of experiments
that go like with an aha moment.
Like you rarely get the Nobel Prize
for like, you've solved everything, we're done.
It's like some inkling of some deep truth.
Like what do you think that would actually look like?
Would it be an experimental result?
I mean, it will have to have some kind of experimental,
maybe validation component.
So what would that look like?
- This is an excellent question.
I wanna, sorry, I'm gonna make a quick point,
which is just a slight tangent.
But you know, like when people ask about the origin of mass
and like looking for the Higgs mechanism and things,
they never are like, we need to find the historical origins
of life in the early,
although those things are related, right?
So this problem of origins of life in the lab,
I think is really important.
But the Higgs is a good example
because you had theory to guide it.
So somehow you need to have an explanatory framework
that can say that we should be looking for these features
and explain why they might be there
and then be able to do the experiment
and demonstrate that it matches with the theory.
But it has to be something that is outside
sort of the paradigm of what we might expect
based on what we know, right?
So this is a really sort of tall order.
And I think,
I mean, I guess the way people would think about it
is like, you know, if you had a bacteria
that climbed out of your test tube or something,
and it was like, you know, moving around on the surface,
that would be ultimate validation.
You saw the original life in an experiment.
But I don't think that's quite what we're looking for.
I think what we're looking for is evidence
of when information that originated
within the bounds of your experiment,
and you can demonstrably prove emerged spontaneously
in your experiment, wasn't put in by you,
actually started to govern the future dynamics
of that system and specify it.
And you could somehow relate those two features directly.
So you know that the program specifying
what's happening in that system
is actually internal to that system.
Like say you have a chemical thing in a box.
- Well, so that's one Nobel Prize winning experiment,
which is like information in some fundamental way
originated within the constraints of the system
without you injecting anything.
But another experiment is you injected something
and got out information.
So like you injected, I don't know,
like some sugar and like something
that doesn't necessarily feel like it should be information.
- Yeah, so actually no, I mean, sugar is information, right?
So part of the argument here is that every physical object
is, well, it's information,
but it's a set of causal histories
and also a set of possible futures.
So there is an experiment that I've talked a lot about
with Lee Cronin, but also with Michael Lachman
and Chris Kempis who are at Santa Fe,
about this idea that sometimes we talk about
as like seeding assembly,
which is you take a high complexity,
like an object that exists in the universe
because of a long causal history,
and you seed it into a system of lower causal history.
And then suddenly you see all of this complexity
being generated.
So I think another validation of the physics would be,
say you engineer an organism
by purposefully introducing something
where you understand the relationship
between the causal history of the organism
and the say very complex chemical set of ingredients
you're adding to it.
And then you can predict the future evolution of that system
to some statistical set of constraints
and possibilities for what it will look like in the future.
You know, I'm a physical structure,
obviously like I'm composed of atoms,
the configuration of them
and the fact that they happen to be me
is because I'm not actually my atoms,
I am a informational pattern
that keeps repatterning those atoms into Sarah.
And I have also associated to me
like a space of possible things that could exist
that I can help mediate come into existence
because of the information in my history.
And so when you understand sort of that
time is a real thing embedded in a physical object,
then it becomes possible to talk about
how histories when they interact,
and a history is not a unique thing,
it's a set of possibilities.
When they interact,
how do they specify what's coming next?
And then where does the novelty come from in that structure?
'Cause some of it is kind of things
that haven't existed in the past can exist in the future.
- Let me ask about this entity that you call Sarah.
- Yes.
I talk to myself, put myself in third person sometimes,
I don't know why.
- So maybe this is a good time to bring up consciousness.
- Sure.
It's been here all along.
- Well, has it?
So that's--
- At least in this conversation,
I think I've been conscious most of it,
but maybe I haven't.
- Well, yeah, so speak for yourself.
You're projecting your consciousness onto me.
You don't know if I'm conscious or not.
- No, I don't.
You're right.
- Is that, you talked about the physics of existence,
you talked about the emergence of causality,
sorry, you talked about causality and time
being fundamental to the universe.
Where does consciousness fit into all of this?
Like, do you draw any kind of inspiration or value
with the idea of panpsychism
that maybe one of the things that we ought to understand
is the physics of consciousness?
Like, one of the missing pieces
in the physics view of the world
is understanding the physics of consciousness.
Or like that word has so many concepts underneath it,
but let's put it,
let's put consciousness as a label on a black box of mystery
that we don't understand.
Do you think that black box holds the key
to finally answering the question of the physics of life?
- The problems are absolutely related.
I think most, and I'm interested in both
because I'm just interested in what we are.
And to me, the most interesting feature
of what we are is our minds
and the way they interact with other minds.
Like minds are the most beautiful thing
that exists in the universe.
So how did they come to be?
- Sorry to interrupt.
So when you say we, you mean humans.
- I mean humans right now,
but that's because I'm a human, or at least I think I am.
- You think there's something special to this particular?
- No, no, no, no, no.
No, I don't, I'm not a human-centric thinker.
- But are you one entity?
You said a bunch of stuff came together to make a Sarah.
Like, do you think of yourself as one entity
or are you just a bunch of different components?
Like, is there any value to understand the physics of Sarah?
Like, or are you just a bunch of different things
that are like a nice little temporary side effect?
- Yeah, you could think of me as a bundle of information
that just became temporarily aggregated into your individual.
Yeah, that's fine.
I agree with that view.
(laughing)
I'll take that as a compliment, actually.
- But nevertheless, that bundle of information
has become conscious,
or at least keeps calling herself conscious.
- Yeah, I think I'm conscious right now,
but I might not be, but that's okay.
Or you wouldn't know.
So yeah, so this is the problem.
So yeah, usually people,
when they are talking about consciousness,
are worried about the subjective experience.
And so I think that's why you're saying,
§07Transcript · 49:04–56:48
I don't know if you're conscious
because I don't know if you're experiencing
this conversation right now.
And nor do you know if I'm experiencing
the conversation right now.
And so this is why this is called
the hard problem of consciousness,
because it seems impenetrable from the outside
to know if something's having a conscious experience.
And I really like the idea of also
the hard problem of matter,
which is related to the hard problem of consciousness,
which is you don't know the intrinsic properties
of an electron not interacting,
say for example, with anything else in the universe.
All the properties of anything that exists in the universe
are defined by its interaction,
because you have to interact with it
in order to be able to observe it.
So we can only actually know the things
that are observable from the outside.
And so this is one of the reasons
that consciousness is hard for science,
because you're asking questions about something
that's subjective and supposed to be intrinsic
to what that thing is as it exists
and how it feels about existing.
And so I have thought a lot about this problem
and its relationship to the problem of life.
And the only thing I can come up with
to try to make that problem scientifically tractable
and also relate it to how I think about the physics of life
is to ask the question,
are there things that can only happen in the universe
because there are physical systems
that have subjective experience?
So does subjective experience have different causes,
things that it can cause to occur
that would happen in the absence of that?
I don't know the answer to that question,
but I think that's a meaningful way
of asking the question of consciousness.
I can't ask if you're having experience right now,
but I can ask if you having experience right now
changes something about you
and the way you interact with the world.
- So does stuff happen?
It's a good question to ask,
does stuff happen if consciousness is--
- Then it's a real physical thing, right?
It has physical consequences.
I'm a physicist, I'm biased, so I don't, you know,
I can't get rid of that bias.
It's really deeply ingrained.
I've tried, but it's hard.
- But I mean, you're saying information is physical too.
So like virtual reality and simulation,
all that program is physical too.
- Yes, everything's physical.
It's just not physical the way it's represented in our minds
- Right, so you, I love your Twitter.
So you tweet these like deep thoughts, deep thoughts.
- That's what a theorist does
when she's trying to experiment.
- Is tweet?
- Yes.
- Sitting there, I mean, I can just imagine
you sitting there for like hours
and all of a sudden just like this thought comes out
and you get a little like inkling into the thought process.
- Yeah, usually it's like when I'm running between things
and I'm like, I've got deep thoughts.
- Well, yeah, so you--
- Deep thoughts are hard to articulate.
- One of the things you tweet is ideologically,
there are many parallels between the search
for neural correlates of consciousness
and for chemical correlates of life.
How the neuroscience and astrobiology communities
treat those correlates is entirely different.
Can you elaborate against this kind of the parallels?
It has to do a little bit with the consciousness
and the matter thing you're talking about.
- Yeah, it does.
And I can't remember what state of mind I was
when I was actually thinking about that.
But I think part of it is--
- I bet you never thought you're gonna have to
analyze your own tweets.
- No, I didn't.
It's an interesting historical juxtaposition of thinking.
So yeah.
- So the tweet is a historical--
- Hey, you're doing an assembly experiment right now
'cause you're bringing a thought from the past
into the present and trying to actually--
- In the lab.
- Yeah, yeah, yeah.
This is experimental science right here
on the podcast live.
- So go, let's see how the consciousness evolves on this one.
- Yeah, so in neuroscience, it's kind of accepted
that we can't get at the subjective aspect of consciousness.
So people are very interested in what would be
a correlate of consciousness.
- What's a correlate?
- A correlate is a feature that relates
to conscious activity.
So for example, a verbal report is a correlate
of consciousness because I can tell you when I'm conscious.
And then when I'm sleeping, for example,
I can't tell you I'm conscious.
So we have this assumption that you're not conscious
when you're sleeping and you're conscious when you're awake.
And so that's sort of like a very obvious example.
But neuroscientists, which I'm no neuroscientist
and I'm not an expert in this field.
But they have very sophisticated ways of measuring activity
in our brain and trying to relate that to verbal report
and other proxies for whether someone
is experiencing something.
And that's what is meant by neural correlates.
And then so when people are trying to think about
studying consciousness or developing theories
for consciousness, they often are trying to build
an experimental bridge to these neural correlates.
Recognizing the fact that a neural correlate
may or may not correspond to consciousness
because that problem's hard
and there's all these associated issues to it.
- So that's from a neuroscience perspective,
it's like fake it till you make it.
So you-- - Pretty much, yeah.
- You fake whatever the correlates are
and hopefully that's going to summon
the thing that is consciousness.
- Yeah, something like that.
- And so the same thing on the chemical correlates of life.
That sounds like, that's an awesome concept.
Is that something that people--
- No, I just made that up.
- Okay.
- That was original to that tweet.
You can cite the tweet.
Maybe I'll write it in a paper someday.
- Chemical correlates of life, that's a good title.
I mean, first of all, your paper is true
that people should check out, have great titles.
- Thank you.
- Papers you're involved with.
So your tweets and titles are stellar
and also your ideas,
but the tweets and titles are much more important.
- Of course.
- So--
- Ideas will live longer.
- Yeah.
- They're much more diffuse though.
- Well, yeah, the tweet is the Trojan horse
for the idea that sticks on for a long time.
Okay, so is there anything to say
about the chemical correlates of life?
You're saying they're similar kind of ways
of thinking about it,
but you mentioned about the communities.
- Yeah, so I think in astrobiology,
there's no concept of chemical correlates of life.
We don't think about it that way.
We think if we find molecules
that are involved in biology, we found life.
So I think one of my motivations there
was just to separate the fact
that life has abstract properties associated to it.
They become imprinted in material substrates
and those substrates are correlates for that thing,
but they are not necessarily the thing
we're actually looking for.
The thing that we're looking for
is the physics that's organizing that system to begin with,
not the particular molecules.
In the same sense that your consciousness
is not your brain.
It's instantiated in your brain.
It has to have a physical substrate,
but the matter is not the thing that you're looking at.
It's some other, at least not in the way
that we have come to look at matter
with traditional physics and things.
There's something else there
and it might be this feature of history I was talking about
or time being actually physically represented there.
- Do you think consciousness can be engineered?
- Yes.
- In the same way that life can be engineered?
- Wow, that was a fast answer.
I didn't even think about that.
That's interesting.
- You don't have a free will.
- No, I do have free will, but it's interesting
'cause I mean--
§08Transcript · 56:49–1:04:21
- Now you're backtracking.
- No, no, I do--
- And that was predestined.
- Yeah, no, no.
No, I do believe in free will,
but I also think that there's kind of an interesting,
speaking about consciousness,
what are you consciously aware of
versus what is your subconscious brain
actually processing and doing?
And sometimes there's conflict
between your consciousness and your subconsciousness
or your consciousness is a little slower
than your subconscious.
And intuition is a really important feature of that.
And so a lot of the ways I do my science
is guided by intuition.
So when I give fast answers like that,
I think it's usually because I haven't really thought
about them and therefore that's probably telling me something.
- Let's continue the deep analysis of your tweets.
You said that determinism in a tweet,
determinism and randomness play important roles
in understanding what life is.
So let me ask on this topic of free will,
what is determinism, what is randomness,
and why the heck do they have anything to do
with understanding life?
- Yeah, and you threw free will in there,
just throwing all the stuff in the bag.
- Are they not related?
Determinism and randomness? - No, no, they are related.
No, no, sorry, I was being unfair.
- You didn't even capitalize the tweet, by the way.
It was all lowercase.
- I must have been angry.
- Oh, that was, can you analyze the emotion behind that?
- No, I actually-- - Is it frustration?
- Yeah, maybe.
So I already argued that I don't think that can happen
without that whole causal history.
And so I guess in some sense,
the determinism for me arises because of the causal history.
And I'm not really sure actually about whether
the universe is random or deterministic.
I just had this sort of intuition for a long time.
I'm not sure if I agree with it anymore,
but it's still kind of lingering,
and I don't know what to do with this question.
But it seems to me, so you asked the question,
what is life?
But you could also, why life?
Why does life exist?
What does the universe need life for?
Not that the universe has needs,
but we have to anthropocentrize things sometimes
to talk about them.
And I had this feeling that if it was possible
for a cup or a desk ornament or a phone on Mars
to spontaneously fluctuate into existence,
the universe didn't need life to create those objects.
It wasn't necessary for their existence.
It was just a random fluke event.
And so somehow to me, it seems that it can't be
that those things form by random processes.
They actually have to have a set of causes
that accrue and form those things,
and they have to have that history.
And so it seems to me that life was somehow deeply related
to the question of whether the underlying rules
of our universe had randomness in them
or they were fully deterministic.
And in some ways, you can think about life
as being the most deterministic part of physics
because it's where the causes are precise in some sense.
- Or most stable?
So like-- - Most stable, yes.
Most reliable.
- Most reliable for the tools of physics.
But what--
- Right, well, so--
- Where's randomness come from then?
Okay, so you were speaking with--
- I've gone in a tangent,
so I'm not sure where we are in the, yeah.
- All of the universe is a kind of tangent,
so we're embracing the tangent.
So free will, you believe-- - Yes.
- At this current time that you have free will--
- I believe my whole life I have free will.
What is illusion?
No, just kidding.
(Lex laughing)
I still believe it.
- You still believe it.
So at the same time, you think that
in your conception of the universe,
causality seems to be pretty fundamental.
- That's right.
- Which kind of wants the universe to be deterministic.
So how the heck--
- Because I'm a determ--
- Do you think you have a free will
and yet you value causality?
- Because I depart from the conception of physics
that you can write down an initial condition
and a fixed law of motion
and that will describe everything.
There's no incompatibility
if you are willing to reject that assertion.
- So where's the randomness,
where's the magic that gives birth to the free will?
Is it the randomness of the laws of physics?
- No, in my mind what free will is
is the fact that I as a physical system
have causal control over certain things.
I don't have causal control over everything,
but I have a certain set of things.
And I'm also, as I described,
sort of a nexus of a particular set of histories
that exist in the universe
and a particular set of futures that might exist.
And those futures that might exist
are in part specified by my physical configuration as me.
And therefore, it may not be free will
in the traditional sense.
I don't even know what people mean
when they're talking about free will, honestly.
It's like the whole discussion's really muddled.
But in the sense that I am a causal agent,
if you wanna call it that, that exists in the universe,
and there are certain things that happen
because I exist as me, then yes, I have free will.
- No, but do you, Sarah,
have a choice about what's going to happen next?
- Oh, I see.
- Could I have, if I run this universe again--
- Yes, I think so.
- You have a choice.
Where's the choice come from?
- I think that's related to the physics of consciousness.
So one of the things I didn't say about that,
I don't know, maybe this is me just being hopeful
because maybe I just wanna have free will.
But I don't think that we can rule out the possibility
because I don't think that we understand enough
about any of these problems.
But I think one of the things that's interesting for me
about the sort of inversion of the question
of consciousness that I proposed
is one of the features that we do
is we have imagination, right?
And people don't think about imagination
as a physical thing, but it is a physical thing.
It exists in the universe, right?
And so I'm really intrigued by the fact that, say,
humans for, you know, another physical system
could do this too, it's not special to humans,
but for centuries imagined flying machines and rockets
and then we finally built them, right?
So they were represented in our minds
and on the pages of things that we drew
for hundreds of years before we could build
those physical objects in the universe.
But certainly the existence of rockets is in part
caused by the fact that we could imagine them.
And so there seems to be this property
that some things don't exist,
they've never physically existed in the universe,
but we can imagine the possibility of them existing
and then cause them to exist,
maybe individually or collectively.
And I think that property is related to what I would say
about having choice or free will,
because that set of possibilities,
that thing, those set of things that you can imagine
is not constrained to your local physical environment
and history.
And this is what's a little bit different
about intelligence as we see it in humans
and AI that we wanna build than biological intelligence
because biological intelligence is predicated completely
on the history of things that's seen in the past.
But something happened with the neural architectures
that evolved in multicellular organisms
that they don't just have access to the past history
of their particular set of events,
but they can imagine things that haven't happened,
aren't on their timeline,
and as long as they're consistent with the laws of physics,
make them happen.
- So this is fascinating.
- It's trippy physics, but it exists, so there you go.
- I mean, in some sense, if you look at like
general relativity and gravity morphing space-time,
in that same way, maybe whatever the physics
of consciousness might be, it might be morphing,
that's like what free will is.
It's morphing like the space,
just like ideas make rockets come to life.
§09Transcript · 1:04:25–1:12:36
It's somehow changing the space of possible realizations
of like whatever's, yeah, okay.
- Life is kind of basically, if you wanna think about it,
like life is sort of changing the probability distributions
over what can exist.
That's the physics of what life is,
and then consciousness is this sort of layered property,
your imagination on top of it,
that kind of scrambles that a little bit more
and like has access to, I don't know.
It's kind of, we don't know how to describe it, right?
Like that's why it's interesting, but--
- But it's probabilistic, so you do think
like God plays dice, so let me--
- No, I think the description's probabilistic.
I don't necessarily think the underlying physics
is probabilistic.
I think the way that we can describe this physics
is going to be probabilistic and statistical,
but the underlying, like when we take measurements
in the lab, but the underlying physics itself
might still be deterministic.
I don't know, maybe I'm, it's hard to know
what concepts to hold onto,
so I find myself constantly rejecting concepts,
but then I have to grab another one
and try to hold onto something from intellectual history.
- Well, it's possible that our mind is not able
to hold the correct concepts in mind at all.
Like we're not able to even conceive of them correctly.
Maybe the words deterministic or random
are not the right even words, concepts to be holding.
But maybe you can talk to the theory of everything,
this attempt in the current set of physical laws
to try to unify them.
Is there any hope that once a theory of everything
is developed, and by theory of everything,
I mean in a narrow sense of unifying quantum field theory
and general relativity, do you think that will contain
some, like in order to do that unification,
you would have to get something that would then give hints
about the physics of life, physics of existence,
physics of consciousness?
- Yeah, I used to not, but I actually,
I have become increasingly convinced that it probably will.
And part of the reason is, I think I've talked
a little bit already about these holes in physics,
like these, the theories we have in physics,
you know, they have problems, they have lots of problems,
and they're very deep problems,
and we don't know how to patch them.
And some of those problems become very evident
when you try to patch quantum mechanics
and general relativity together.
So there is this kind of interesting feature
that some of the ways of patching that
might actually closely resemble the physics of life.
And so the place where that actually comes up most,
and actually we just had a workshop in the Beyond Center
where I work at Arizona State University,
and Lee Smolin made this point that he thinks
that the theory of quantum gravity, when we solve it,
is gonna be the same theory that gives rise to life.
And I think that I agree with him on some levels
because there's something very interesting
where if you look at these sort of causal set theories
of gravity, where they're looking for space
as being emergent, and so space-time is an emergent concept
from a causal set, which is also sort of related,
I think, to what Wolfram's doing with his physics project.
It's the same kind of underlying math
that we have in this theory that we've been developing
related to life called assembly theory,
which is basically trying to look at complex objects
like molecules and bacteria and living things
as basically being assembled from a set of component parts
and that they actually encode all the possible histories
that they could have in that physical object.
So mathematically, all these ideas, I think,
are related.
I think a lot of people are thinking about this
from different perspectives.
And then constructor theory that David Deutsch
and Chiara Marleto have been developing
is a totally different angle on it,
but I think getting at some similar ideas.
So it's a really interesting time right now, I think,
for the frontiers of physics and how it's relating
to maybe deeper principles about what life is.
So short answer, yes.
Long-winded answer, rewind.
- Can we talk about aliens?
- Anytime.
(laughing)
- So one, I think one interesting way
to sneak up on the question of what is life
is to ask what should we look for in alien life?
If we were to look out into our galaxy
and into the universe and come up with a framework
of how to detect alien life,
what should we be looking for?
Is there a set of rules?
It's both the tools and the tools that are,
serve as sensors for certain kind of properties of life.
So what should we look for in alien life?
- Yeah, so we have a paper actually coming out on Monday,
which is collaboration.
It's actually really Lee Cronin's lab,
but my group worked with him on it
and we're working on the theory,
which is this idea that we should look for life
as high assembly objects.
What we mean by that is,
which is actually observationally measurable.
And this is one of the reasons
that I started working with Lee on these ideas
is because being a theorist, it's easy to work in a vacuum.
It's very hard to connect abstract ideas
about the nature of life
to anything that's experimentally tractable.
But what his lab has been able to do
is develop this method where they look at a molecule
and they break it apart into all its component parts.
And so you say you have some elementary building blocks
and you can build up all the ways of putting those together
to make the original object.
And then you look for the shortest path in that space.
And you say that's sort of the assembly number
associated to that object.
And if that number's higher,
it assumes that a longer causal history
is necessary to produce that object
or more information is necessary
to specify the creation of that object in the universe.
Now that kind of idea at a superficial level
has existed for a long time.
That kind of idea as a physical observable of molecules
is completely novel.
And what his lab has been able to show
is that if you look at a bunch of samples
of non-biological things and biological things,
there's this kind of threshold of assembly
where as far as the experimental evidence is
and also your intuition would suggest
that non-biological systems don't produce things
with high assembly number.
So this goes back to the idea
like a protein's not gonna spontaneously fluctuate
into existence on the surface of Mars.
It requires an evolutionary process
and a biological architecture to produce a protein.
You generalize that argument,
you know a complex molecule or a cup or a desk ornament
in this sort of abstract idea of assembly spaces
as being the causal history of objects.
And you can talk about the shortest path
from elementary objects to an object
given an elementary set of operations.
And you can experimentally measure that with a mass spec.
And that's basically sort of the idea.
- That's really fascinating.
I can't get out of my head.
I'd start imagining Legos
and all the Legos I've ever built and how many steps.
What is the shortest path to the final--
- Right, right.
- To find a little Lego castles.
- So yeah, so then like asking
about going to look for alien life,
the idea is most of the instruments that NASA builds,
for example, or any of the space agencies
looking for life in the universe
are looking for chemical correlates of life, right?
But here we have something
that is based on properties of molecules.
It's not a chemical correlate.
It's agnostic.
It doesn't care about the molecule.
It cares about what is the history necessary
to produce this molecule.
How complex is it in terms of how much time is needing,
how much information is required to produce it.
- So when you observe a thing on another planet,
you're essentially,
the process looks like reverse engineering,
trying to figure out what is the shortest path
to create that thing.
- Yeah, so most, yeah.
And I would say most,
like most examples of biology or technology
don't take the shortest path, right?
But the shortest path is a bound on how hard it is
for the universe to make that.
- Yeah, and I guess you and Lee are saying
that there's a heuristic,
that's a good metric for,
like better perhaps than chemical correlates.
- Yes, because it doesn't,
it's not contingent on looking for the chemistry of life
on Earth, on other planets.
§10Transcript · 1:12:38–1:20:18
And it also has a deeper explanatory framework
associated to it, as far as the kind of theory
that we're trying to develop
associated to what life is.
And I think this is one of the problems I have
in my field personally in astrobiology,
is people observe something on Earth,
say oxygen in the atmosphere,
or an amino acid in a cell,
and then they say,
let's go look for that on another planet.
Let's look for oxygen on exoplanets,
or let's look for amino acids on Mars.
And then they assume that's a way of looking for life.
Or even phosphine on Venus.
But you know, like there's all these examples
of let's look for one molecule.
A molecule is not life.
Life is a system that patterns
particular structures into matter.
That's like, that's what it is.
And it doesn't care what molecules are there.
It's something about the patterns
and that structure and that history.
And if you're looking for a molecule,
you're not testing any hypotheses
about the nature of what life is.
It doesn't tell me anything
if we discover oxygen on an exoplanet
about what kind of life is there.
Just oxygen on an exoplanet.
It's not, there's,
I guess I think like when you think about the question,
are we alone in the universe?
That's a pretty fricking deep question.
It should have a fricking deep answer.
It shouldn't just be there's a molecule on an exoplanet.
Wow, we solved the problem.
It should tell us something meaningful about our existence.
And I feel like we've fallen short
on how we're searching for life
in terms of actually searching for things like us
in this kind of deeper way.
- But how do you do that initial kind of,
say I'm walking down the street
and I'm looking for that double take test of like,
like what the hell is that?
Like that initial,
like how do we look for the possibility of weirdness
or the possibility of high assembly number?
What would aliens look like
if they don't have two eyes and are green?
- If I knew, I would have probably already solved the problem.
- Right, there's another Nobel prize in there somewhere.
- Yeah, somewhere in there.
Well, I think it's kind of,
so there is a bias here, right?
So we've evolved to recognize life on earth, right?
Like I, you know, children at a very early age
can tell the difference between a puppy and a plant
and then the plant and a chair, for example,
you know, like it just, it seems innate.
And so I think, and also because we're life,
you know, I think like there's this implicit bias
that we should know it when we see it
and it should be completely obvious to us.
But there are a lot of features of our universe
that are not completely obvious to us.
Like the fact that this table is made of atoms
and that I'm sitting in a gravitational
potential well right now.
And I guess my point with this is,
I think life is much less obvious than we think it is.
And so it could be in many more forms than we think it is.
And I guess this goes back to the point
about being open-minded that we may not know
what alien life looks like.
It might not even be possible to interact with alien life
'cause maybe something about, you know,
our informational lineage, it makes it impossible
for information from an alien to be copied to us.
Therefore there's no, you know,
so to speak communication channel.
And I don't mean, you know, verbal communication,
just it's not in our observational space.
Like, you know, like, you know,
there's fundamental questions about why we observe
the universe in position rather than momentum,
but we also, you know, observe it
in terms of certain informational patterns and things.
Like that's what our brain constructs
and maybe aliens just interact
with a different part of reality than we do.
That's wildly speculative, but I think--
- But it's possible.
- It's possible and I think it's consistent with the physics.
So I think the best ways we can ask questions
are about life and chemistry and asking questions
about if information is a real physical thing,
what would its signatures be in matter?
And how do we recognize those?
And I think the ones that are most obvious
are the ones I've already articulated.
You have these objects that seem completely improbable
for the universe to produce
because the universe doesn't have the design
of that object in the laws.
So therefore, an object had to evolve.
We talk, we call it evolution,
but it had to be produced by the universe
that then had all of the possible tasks
to make that object specified.
- I mean, there's some, like,
there's an engineering question here of,
are there sensors we can create that can give us,
can help us discover certain pockets
of high assemblies, aliens?
Like, I mean, there is a hope,
setting dogs and chairs aside,
there's a hope that visually we could detect.
Like, because our universe,
I mean, at least the way we look at it now,
like this three-dimensional, like space-time,
we can visually comprehend it.
It's interesting to think, like, if we got to hang out,
you know, if there's an alien in this room,
like, would we be able to detect it
with our current sensors?
Not the fancy kinds, but like webcams.
- Like, say, standing over there?
- Yeah, standing over there,
or maybe like in this carpet.
See, there's all these kinds of patterns, right?
- Yeah.
- I don't know if this carpet is an alien.
- Well, so I see what you're saying.
So assembly theory is pretty general.
Like, I mean, we've been applying it to molecules
because it makes sense to apply it to molecules,
but it's supposed to explain life,
you know, like the physics of life.
So it should explain, you know, the things in this room
in addition to molecules.
So I guess, and you can apply it to images and things.
So I guess the idea, you know, you could explore
is just looking at everything on planet Earth
in terms of its assembly structure,
and then looking for things
that aren't part of our biological lineage.
If they have high assembly, they might be aliens on Earth.
- I mean, that is a very kind of
rigorous computer vision question.
Can we visually, is there a strong correlation
between certain kind of high assembly objects
when they get to the scale where they're visually observable
and some, like when it's, say, projected onto a 2D plane,
can we figure out something?
- I'm glad you brought up the computer vision point
'cause for a while I had this kind of thought in my mind
that we can't even see ourselves clearly.
So one of the things, you know,
people are worried about artificial intelligence
for a lot of reasons, but I think it's really fascinating
'cause it's like the first time in history
that we're building a system
that can help us understand ourselves.
So like, you know, people talk about AI physics,
but like, you know, when I look at another person,
I don't see them as a four billion year lineage,
but that's what they are, and so is everything here, right?
So imagine that we built artificial systems
that could actually see that feature of us.
What else would they see?
And I think that's what you're asking.
And I think that would be so cool.
(both laughing)
I want that to happen,
but I think we're a little ways off from it,
but yeah, we're going there, I hope.
- Okay, let me ask you, I apologize ahead of time,
but let me ask you the internet question.
So you're a physicist, you ask rigorous questions
about the physics of existence
and these models of high assembly objects.
Now, when the internet would see an alien,
they would ask two questions.
One, can I eat it?
And two, can I have sex with it?
- Yes.
- So... (laughs)
- All the existential questions.
Those are very important.
- The internet is very sophisticated.
- It really is.
It's gotten our basal cognition pretty good.
- So you kind of mentioned that it's very difficult.
It's possible that we may not be even able
to communicate with it.
- Right, I think the internet has more hope than we do.
- Yeah, it's a hopeful place, yes.
§11Transcript · 1:20:20–1:28:23
Do you think in terms of like interacting
on this very primal level of sharing resources,
like what would aliens eat?
What would we eat?
Would we eat the same thing?
Could we potentially eat each other?
One person eats the other or the aliens eat us.
And the same thing with not sex in general reproduction,
but genetically mixing stuff.
Like would we be able to mix genetic information?
- Maybe not genetic, but maybe information, right?
And I think part of your question is like,
so if you think of life as like this history
of events that happen in the universe,
like there's this question of like,
how divergent are those histories, right?
So when we get to the scale of technology,
it's possible to imagine,
imagine, although we can't even do it,
like imagine all the possible technologies
that could exist in the universe.
But if you think about all the possible chemistries,
somehow that seems like a lower dimensional space
and a lower set of possibilities.
So it might be that like when we interact with aliens,
we do have to go back to those more basal levels
to figure out sort of what the map is, right?
Like the sort of where we have a common history.
We must have a common history somewhere in the universe,
but in order to be able to actually interact
in a meaningful way, you have to have some shared history.
I mean, the reason we can exchange genetic information
and eat each other's food or eat each other as food
is because we have a shared history.
- So we have to find that shared history.
We have to find the common ancestor
in this causality map, this causality tree.
- Yes, and we have a last universal common ancestor
for all life on earth, which I think is sort of the nexus
of that causality map for life on earth.
But the question is, where would other aliens diverge
on that map? - That's really interesting.
So say there's a lot of aliens out there in the universe,
each set of organisms would probably have like a number,
you know, like Erdos number of like how far,
like how far our common ancestor is.
And so the closer the common ancestor, like it is on earth,
the more like each other, the more likely we are
to be able to have sexual reproduction.
- Well, it's like sort of like humans
having common culture and languages, right?
- Yeah, exactly, language, communication.
- It might take a lot of work though with an alien
'cause you really have to get over a language barrier.
- Oh boy, so it's communication, it's resources.
I mean, it's the whole.
- And I think tied into that is the questions
of like who's gonna harm who.
And actually definitions of harm.
- And whether your parents approve,
you know, all those kind of questions.
- Whether the common ancestor approves, yeah.
It's just very true.
How many alien civilizations do you think are out there?
- I don't have intuition for that,
which I have always thought was deeply intriguing.
So, and part of this, I mean, I say it specifically
as I don't have intuition for that
because it's like one of those questions
that you feel around for a while
and you really just, you can't see it,
even though it might be right there.
And in that sense, it's a little like
the quantum to classical transition.
You're like really talking about
two different kinds of physics.
And I think that's kind of part of the problem.
Once we understand the physics,
that question might become more meaningful.
But there's also this other issue.
And this was really instilled on me
by my mentor, Paul Davies, when I was a postdoc
'cause he always talks about how, you know,
whether aliens are common or rare is kind of just,
you know, it follows a wave of popularity
and it just depends on like the mood of, you know,
what the culture is at the time.
And I always thought that was kind of
an intriguing observation,
but also there's this set of points about
if you go by the observational evidence,
which we're supposed to do as scientists, right?
You know, we have evidence of us
and one original life event from which we emerged.
And people wanna make arguments
that because that event was rapid
or because there's other planets
that have properties similar to ours,
that that event should be common.
But you actually can't reason on that
because our existence observing that event
is contingent on that event happening,
which means it could have been completely improbable
or very common.
And Brandon Carter like clearly articulated that
in terms of anthropic arguments a few decades ago.
So there is this kind of issue
that we have to contend with dealing with life
that's closer to home than we have to deal with
with any other problems in physics,
which we're talking about the physics of ourselves.
And when you're asking about the original life event,
that event happening in the universe,
at least it's like our existence is contingent on it.
And so you can think about sort of fine tuning arguments
that way too.
So, but the sort of other part of it is like,
when I think about how likely it is,
I think it's because we don't understand this mechanism yet
about how information can be generated spontaneously.
That I like, 'cause I can't see that physics clearly yet,
even though I have a lot of,
like some things around the space of it in my mind,
I can't articulate how likely that process is.
So my honest answer is, I don't know.
And sometimes that feels like a cop out,
but I feel like that's a more honest answer
and a more meaningful way of making progress
than what a lot of people wanna do, which is say,
oh, well, we have a one in 10 chance of having it
on an exoplanet with Earth-like properties
because there's lots of Earth-like planets out there
and life happened fast on Earth.
- Well, so, kind of a follow-up question,
but as a side comment,
what I really am enjoying about the way
you're talking about human beings is you always say,
and not to make yourself conscious about it,
'cause I really, really enjoy it, that you say we.
- Yes. - You don't say humans.
You say, 'cause oftentimes, like, you know,
I don't know, evolutionary biologists
will kind of put yourself out as an observer,
but it's kind of fascinating to think
that you as a human are struggling about your own origins.
- Yes, that's the problem, and yeah,
and I think, I don't do that deliberately,
but I do think that way,
and this is sort of the inversion
from the logic of physics,
because physics, as it's always been constructed,
has treated us as external observers of the universe,
and we are not part of the universe,
and this is why the problem of life, I think,
demands completely new thinking,
because we have to think about ourselves
as minds that exist in the universe,
and are at this particular moment in history,
and looking out at the things around us,
and trying to understand what we are inside the system,
not outside the system.
We don't have descriptions at a fundamental level
that describe us as inside the system,
and this was my problem with cellular automata also.
You're always an external observer for a cellular automata.
You're not in the system.
What does a cellular automata look like from the inside?
- I think you just broke my brain with that question.
- Exactly, but that's the fundamental--
- I thought about that for a long time, but.
(laughing)
- Yeah, that's a really clean formulation
of a very fundamental question,
'cause you can only, to understand cellular automata,
you have to be inside of it,
but as a human, sort of a poetic, romantic question,
does it make you sad?
Does it make you hopeful, whether we're alone or not?
In the different possible versions of that,
if we're the highest assembly object in the entire universe,
does that give you--
- At this moment in time, maybe.
- At this moment in the causality--
- 'Cause we may, I assume we have a future.
- Well, we definitely have a future.
The question is where that future decreases the assembly.
It could be we're at the peak, or we could be just--
- That would be inconsistent with the physics in my mind,
but, so I should give a caveat.
I've given the caveat that I'm biased as a physicist,
but I'm also biased as an eternal optimist,
so pretty much all of my modes of operation
for building theories about the world
are not like an Occam's razor,
what's the simplest explanation,
but what's the most optimistic explanation?
And part of the reason for that is if you really think
explanations have causal power,
in the sense that the fact that we have theories
about the world has enabled technologies
§12Transcript · 1:28:24–1:35:55
and physically transformed the world around us.
I think I have to take seriously that
as a part of the physics I wanna describe,
and try to build theories of reality
that are optimistic about what's coming next,
because the theories are in part
the causes of what comes next.
- So there could be a physics of hope
or a physics of optimism in there too.
- Yes.
- 'Cause that seems like also, I mean,
optimism does seem to be a kind of engine
that results in innovation.
- Yes.
- So this is, like, why the hell
are we trying to come up with new stuff?
- Oh, so I made this point about thinking
life is the physics of existence,
and it's not just the physics of existence,
it's the physics of more things existing.
So I think one of these drives--
- Creativity.
- Yeah, creativity, like optimism.
People like entropy, I don't like entropy
as it was formulated in the 1800s,
I think it's an antiquated concept,
but this idea of maximizing over the possible number
of states that could exist.
Imagine the universe is actually trying to maximize
over the number of things that could physically exist.
What would be the best way to do that?
The best way to do that would be
evolve intelligent technological things
that could explore that space.
- So okay, that's talking about alien life
out there in the universe,
but you've also earlier in the conversation
mentioned the shadow biosphere.
So is it possible?
That we have weird life here on Earth
that we're just not,
like even in a high assembly formulation of life,
that we're just not paying attention to, we're blind to?
Like life we're potentially able to detect,
but we're blind to?
And maybe you could say, what is the shadow biosphere?
- Sure, sure.
Yeah, the shadow biosphere is this idea
that there might have been other original life events
that happened on Earth that were independent
from the original life event that led to us
and all of the life that we know on Earth.
And therefore there could be aliens
in the sense they have a different origin event
living among us.
And it was proposed by a number of people,
but one of them was Paul Davies
that I mentioned earlier as my mentor.
And he has a really cute way of saying
that aliens could be right under our noses
or even in our noses.
With a British accent, it sounds better.
But anyway, so the idea is like,
it could literally be anywhere around us.
And if you think actually about the discovery
of like viruses and bacteria,
for a long time, they were kind of a shadow biosphere.
It was life that was around us, but invisible.
But this takes it a little bit further
in saying that all of those examples, viruses, bacteria,
and everything that we've discovered so far
has this common ancestry
and the last universal common ancestor of life on Earth.
So maybe there was a different origin event
and that life is weirder still and might be among us
and we could find it.
We don't have to go out and the stars look for aliens
just here on Earth.
- Do you think that's a serious possibility
that we should explore with the tools of science?
Like this should be a serious effort?
- I think yes and no.
And I mean, yes, because I think it's a serious hypothesis
and I think it's worth exploring
and it is certainly more economical
to look for signs of alien life on Earth
than it is to go and build spacecraft
and send robots to other planets.
And that was one of the reasons it was proposed is,
well, if we do find an example
of another original life on Earth,
it's hugely informative
because it means the original life is not a rare event.
If it happened twice on the same planet,
that means it's probably pretty probable
given conditions are right.
So it has huge potential scientific impact,
not to mention the fact that you might have
like biochemistry and stuff that's informative
for like medicine and stuff like that.
But I think that the thing for me
that's challenging about it,
and this really comes from my own work,
like thinking about life as a planetary scale process
and also trying to understand
sometimes what I call like
the statistical mechanics of biochemistry,
but large scale statistical patterns
in the chemistry that life uses on Earth.
There are a lot of regularities there
and life does seem to have planetary scale organization
that's consistent even with some of the patterns
that we see at the individual scale.
So if you think life is a planetary scale phenomena
and the chemistry of life has to be sort of,
not just, it's not,
an individual is not necessarily
the fundamental unit of life, right?
The fundamental unit of life
is these informational lineages
and they're kind of,
they intersect over spatial scales.
So everything on Earth is kind of related
by the common causal history.
So it's hard for me,
based on the way I think about the physics
and also some of the stuff that my group has done,
to really think that there could be evidence
or there could be a second sample of life on Earth.
But I think there are ways
that we need to be more concrete about that.
And I have thought a little bit about,
like you can represent the chemistry
in an individual cell as a network.
And then those networks,
something my group has shown,
actually scale with the same property.
So ecosystems have the same properties
as individuals as planetary scale.
And then you could imagine
if you had alien chemistry intermixed in there,
that scaling would be broken.
So if there's some robustness property
or something associated to it,
and you get alien chemistry in there,
it just breaks everything.
And you don't have a planetary ecosystem functioning
and individuals functioning across all these scales.
So I guess what I'm arguing
is life is not a scale dependent phenomena.
It's not just cellular life.
So if you have a shadow biosphere,
it has to be integrated with all of these other scales.
- And that would lose the meaning
of the word shadow biosphere, I guess.
- I think so, yeah.
So it's an open question, right?
And I think it would tell us a lot.
So there has been very minimal effort
of people to look for a shadow biosphere.
- But then the question,
it could be possible that there's like sufficiently
distinct planets within one planet,
meaning like environments within one planet.
Like, I don't know.
I've been looking recently
because of having a chat with Catherine DeCleer
about Io, the moon of Jupiter,
that's like all volcanoes and volcanoes are bad-ass.
But like imagining-
- Io's bad-ass.
- Imagining life inside volcanoes, right?
It seems like sufficiently chemically different
like to be living in the darkness
where there's a lot of heat
and maybe you can have different Earths on a planet.
- Yeah, or like if you go deep enough in the crust,
maybe there's like a layer where there's no life
and then there's suddenly life again.
And maybe those lizard men or whatever
that people dream about are really down there.
I know that's a little flippant,
but really like there could be like chemical cycles
deep in the Earth's crust that might be alive
and are completely distinct in chemical origin
to surface life.
- Right, that they wouldn't be interacting with each other.
- Yeah, and that's one of the proposals
for the shadow biosphere is like,
sometimes people talk about it as being geologically
or geographically distinct that it might be,
you have no life for this region
and then a different example.
And then sometimes people talk about it
being chemically distinct,
that the chemistry is sufficiently different,
that it's completely orthogonal
or non-interacting with our chemistry.
- It seems to me at least the chemistry
is a more powerful boundary than geographic.
§13Transcript · 1:36:02–1:44:48
It just seems like life finds a way literally to travel.
What do you think about all these UFO sightings?
So to me, it's really inspiring.
It's yet another localized way to dream
about the mysterious that is out there.
- Yeah, so I've actually been more intrigued
by the cultural phenomena UFOs
than the phenomena UFOs themselves,
because I think it's intriguing about how
we are preparing ourselves mentally
for understanding others
and how we have thought about that historically
and what the sort of modern incarnations of that are.
It's more like, I want an explanation for us,
that's my motivation.
And having some streaks across the sky or something
and saying that's aliens, it doesn't tell you anything.
So unless you have a deeper explanation
and you have more lines of,
where is this gonna take us in the future?
It's just not as interesting to me
as the problem of understanding life itself
and aliens as a more general phenomenon.
- I do think it's just, as you said,
a good way to psychologically and sociologically
prepare ourselves to sort of like,
what would that look like?
And very importantly,
which is what a lot of people talk about politically,
sort of there's this idea from the,
so I came from the Soviet Union of like the Cold War
and we have to hide secrets.
Some way in us searching for life on other planets
or searching for life in general,
the way we've done government in the past,
we tend to think of all new things
as potential military secrets, so we want to hide them.
And one of the ways that people kind of look
at UFO sightings is like,
maybe we shouldn't hide this stuff.
Like what is the government hiding?
I think that's a really,
in one sense it's a conspiratorial question,
but I think in another,
it's an inspiration to change the way we do government
to where secrets don't,
maybe there are times when you want to keep secrets
as military secrets,
but maybe we need to release a lot more stuff
and see us as a human species
as together in this whole search.
- Yeah, the public engagement part there
is really interesting.
- And it's almost like a challenge
to the way we've done stuff in the past
in terms of keeping secrets.
When they're not, so like the first step,
if you don't know how something works,
if there's a mysterious thing,
the first instinct should not be like, let's hide it.
Let's put it in the closet.
So that the Chinese or the Russian government
or whatever government doesn't find it.
Maybe the first instinct should be, let's understand it.
Perhaps let's understand it together.
- No, I think that's good.
And something I realized recently
that I never thought was gonna be a problem,
but I think this actually helps with quite a bit
is because so many people nowadays
believe we've already made contact
that as an astrobiologist,
if we actually wanna understand life and make contact,
we kind of have to deconstruct the narratives
we've already built from ourselves
and kind of unteach ourselves
that we've learned about aliens and then reteach ourselves.
So there's this really interesting sort of dialogue there
and making it open to the public
that they actually have to think critically about it
and they see the evidence for themselves,
I think is really important for that process.
- Yeah, the reteaching,
that aliens might be way weirder than we can imagine.
- Yes, yes.
I'm pretty sure they're probably weirder
than we can imagine.
- Okay, we've in 2020 and still living through a pandemic,
setting the political and all those kinds of things aside,
I've always found viruses fascinating as dynamical systems.
I was gonna say living systems,
but I've always kind of thought of them as living,
but that's a whole nother kind of discussion.
Maybe it'd be great to put that on the table.
One, do you find viruses beautiful/terrifying?
And two, do you think they're living things?
Or there's some aspect to them per our discussion of life
that makes them living?
- I mean, living in a pandemic,
saying viruses are beautiful is probably a hard thing,
but I do find them beautiful to a degree.
I think even in the sense of mediating a global pandemic,
there's something like deeply intriguing there
because these are tiny, tiny little things, right?
And yet they can essentially,
like cause a seizure,
like handicap an entire civilization at a global scale.
So just that intersection between
our perceived invincibility and our susceptibility to things
and also the interaction across scales of those things
is just a really amazing feature of our world.
- Most technology, whether it's viruses or AI,
that can scale in an exponential way,
like kind of run,
as opposed to like one thing makes another thing,
makes another thing,
it's one thing makes two things
and those two things make four things.
Like that kind of process
also seems to be fundamental to life.
- Yes.
- And it's terrifying because in a matter of,
in a very short timescale,
if it's good at being life, whatever that is,
it can quickly overtake the other competing forms of life.
- Right.
- And that's scary both for AI and for viruses.
And it seems like understanding these processes
that are underlying viruses.
And I don't mean like on the virology or biology side,
but on some kind of more computational physics perspective,
as we've been talking about,
seems to be really important
to figure out how humans can survive.
- Right.
- Along with these kinds of,
well, this kind of life
and perhaps becoming a multi-planetary species
is a part of that.
Like there's no,
maybe like we'll figure out from a physics perspective
is like, there's no way any living system
can be stable for a prolonged period of time
and survive unless it expands exponentially throughout.
Like we have to multiply.
Otherwise, anything that doesn't multiply exponentially
will die eventually.
Maybe that's a fundamental law.
- Maybe, I don't know.
I always get really bothered by these Darwinian narratives
that are like the fittest replicator wins and things.
And I just don't feel like that's exactly what's going on.
I think like the copying of information
is sort of ancillary to this other process of creativity.
Right, so like the drive is actually,
the drive is creativity,
but if you wanna keep the creativity
that's existed in the past,
it has to be copied into the future.
So replication, like if you,
so that for me is,
so I had this set of arguments
with Michael Lachman and Lee Cronin
about the like life being about persistence.
They thought it was about persistence
and like survival of the fittest kind of thing.
And I'm like, no, it's about existence.
It's like, 'cause when you're talking about that,
it's easy to say that in retrospect,
you can post-select on the things that survived
and then say why they survived,
but you can't do that going forward.
- That's really profound.
That survival is just a nice little side effect feature
of maximizing creativity,
but it doesn't need to be there.
- Yeah.
- That's really beautiful. - I like that, yeah.
- Yeah, that's really-- - Like I said,
I like optimistic theories.
- Well, I don't know if that's optimistic.
That could be terrifying to people because,
because a system that maximizes creativity
may very quickly get rid of humans
for some reason if it comes up with some other creative,
I mean, forms of existence.
- Yeah.
- Right, this is the AI thing.
It's like the moment you have an AI system
that can flourish in the space of ideas
or in some other space much more effectively than humans,
and it's sufficiently integrated into the physical space
to be able to modify the environment.
- I think we'll just be like
the core genetic architecture or something.
We'll be like the DNA for AI, right?
- Yeah. - It's like we haven't lost
the past informational architectures on this planet.
They're still there.
- Yeah, so the AI will use our brains in some part
§14Transcript · 1:44:53–1:53:30
to like ride, like it'll accelerate the exchange of ideas.
That's the neural language dream is that,
well, the humans will be still around
'cause you're saying architecture will still--
- Yeah, but I don't even think they necessarily
need to tap in our brains.
I mean, just collectively, we do interesting things.
What if they were just using like the patterns
in our communication or something?
- Oh, without controlling it, just observing?
- Well, I don't know.
In what sense do you control the chemistry
happening in your body?
- Hmm, yeah.
- I mean, obviously I don't know.
People look at AI and then they look at this thing
that's bigger than us and is coming in the future
and is smarter than us, and I think though
that looking at the past history of life on the planet
and what information has been doing
for the last four billion years
is probably very informative to asking questions
about what's coming next.
One is planetary scale transitions are really important
for new phases, so the global internet
and sort of global integration of our technology
I think is an important thing.
So that's again, life as a planetary scale phenomena,
but we're an integrated component of that phenomena.
I don't really see that the technology
is gonna replace us in that way.
It's just gonna keep scaffolding and building.
And I also don't have an idea
that we're gonna build AI in a box.
I think AI is gonna emerge.
AGI to me is a planetary scale phenomena
that's gonna emerge from our technology.
- Planetary scale phenomena.
But do you think, and AGI is not distinct from humans.
The whole package-- - The whole package, yeah.
- Comes as a planetary scale phenomena.
- And that goes back to the fact that
you were asking questions about you as an individual.
Like what are you as an individual?
You're like a packet of information
that exists in the particular physical thing that is you.
We're all just packets of information
and some of us are aggregates in certain ways,
but it's all just kind of exchanging
and propagating, right, and processing.
- Is your packet of information
that you've continually referred to as Sarah
afraid of the dissipation of the death of that packet?
Are you afraid of death?
Do you ponder death?
Does death have meaning in this process of creativity?
- I think I have the natural biological urge
that everyone has to fear death.
I think the thing that I think is interesting
is if I think about it rationally,
I'm not necessarily afraid of death for me
because I won't be aware of being dead.
But I am afraid for my kids
because it matters to them if I die.
So again, I think death becomes more significant
as a collective property, not as an individual one.
- Yeah, but isn't there something to fear
about the fact that the way,
like the creative,
the complexity of information
that's been created in you,
the fact that it kind of breaks apart and disappears.
- It doesn't, but I don't think it disappears.
It's just not me anymore.
- Right, but that process of it being not you anymore,
that doesn't scare you?
- Of course it does.
- The mystery of it.
I mean the-- - Yeah.
But I guess I'm heartened by the fact
that there will be some imprints of the fact
that I existed still in the universe after I leave it.
- Yeah, but there'll be a, okay.
- And also that has to do with my perception of time, right?
So I perceive time as flowing,
but that might not be the case.
I mean, this is standard physicist comfort
is every moment exists,
and the flow of time is just our perception of us changing.
- So you can travel back in time and that's comforting?
Like from a physicist's--
- No, no, no, I'm not talking about traveling back in time.
I'm just saying that the moments in the past still exist.
Now whether the moments in the future exist or not
is a different question.
- That's not comforting to me in terms of death.
The flow of time is not, that does not--
- I think there's no comfort in the face of death
for what we are because we like existing.
And I think it's especially true if you love life
and you love what life is.
- Do you think there's a certain sense in which
the fear of death or the fear of non-existence,
maybe fear is not the right word,
is the actual very phenomena that gives birth to existence?
Like death is fundamental.
Like it just feels like freaking out,
oh shit, this ride ends, is actually like,
that's the thing that gives birth to this whole thing.
It's constantly, it's matter constantly freaking out
about the fact that it's gonna be--
- No, I think things like to exist.
I think they wanna exist.
- Yeah, there's a desire or whatever to exist.
- Yeah, there's a drive to exist
and there's a drive for more things to exist.
I guess, yeah, I like existing.
I like it a lot.
And I don't know it any other way.
- See, I don't even know if I like existing.
I think I really don't like not existing.
- Yes, yeah, that too.
(laughing)
Yeah, maybe it's that.
Some days I might like existing less than others.
(laughing)
- Yes, but I think those are like surface feelings.
There is some, seems like there's something fundamental
about wanting to exist.
- No, I think that's right, but I think to your point
that that might go back to the more fundamental idea
that if life is the physics of existence
and maximizing existence, individual organisms,
of course, wanna maximize their existence
and everything wants to exist.
But I guess for me, the small comfort is
my existence matters to future existence.
- Speaking of future existence,
is there advice you can give to future pockets of existences
aka young people about life?
You've had, you've worn many hats,
you've taken on some of the biggest problems in the universe.
Is there advice you can give to young people about life,
about career, about existing?
- Maybe not about the last one.
A lot of people ask me this question
about like working on such hard problems,
like how can you make a successful career out of that?
But I think for me, it couldn't be otherwise.
Like I have to, to be fulfilled,
you have to work on things you care about
and that's always kind of driven me.
And that's been discipline, department
and sort of superficial level problem independent
because I started at community college actually
and I was taking a physics class
and I learned about magnetic monopoles
and we didn't know if they existed in the universe
but we could predict them and we could go look for them.
And I was so deeply intrigued by this idea
that we had this mathematical formula
to go look for things.
And then I wanted to become a theoretical physicist
because of that,
but that actually wasn't my driving question.
I realized my driving question is the nature
of the correspondence between our minds
and physical reality and what we are.
And that question is very deep
so you can work across a lot of fields doing that.
But I think without that driving question,
I never would have been able to do all the things
that I've done.
It's really the passion that drives it.
And usually when students ask me these kinds of questions,
I tell them like,
you have to find something you really care about working on
because if you don't really care about it,
A, you're not gonna be your best at it
and B, it's not gonna be worth your time.
Why would you spend your time working on something
you're not interested in?
- So find the driving questions.
- Yeah, find the driving question, find your passion.
I mean, I think passion makes a huge difference
in terms of creativity, talent and potential
and also being able to tolerate all the hard things
that come with any career or life.
- Yeah, I've had a bunch of moments in my life
where I've just been captivated by some beautiful phenomena
and I guess being rigorous about it
and asking what is the question underlying this phenomenon
like robots bring a smile to my face
and forming a question of like,
why the hell is this so fascinating?
Why is this specifically
the human robot interaction question
that something beautiful is brought to life
when humans and robots interact,
understanding that deeply.
§15Transcript · 1:53:32–1:59:21
I was like, okay, so this is gonna be my life work then.
I don't know what the hell it is,
but that's what I wanna do.
And doing that for whatever the hell gives you
that kind of feeling, I guess is the point.
Am I allowed to ask you a question?
- Sure.
- Okay, on that point,
'cause I had this colleague that suggests the idea
that consciousness might be contagious
and so interacting with things.
- That's funny.
No, yeah, yeah, it's a beautiful way of putting it.
- It's an interesting idea, right?
So I'm wondering the motivation there.
Is it the motivation that you want more of the universe
to appreciate things the way we do
and appreciate those interactions
or is it really more the enjoyment of the human
in those interactions?
Is it, do you know what I'm asking?
- Yeah, yeah, yeah.
See, I think consciousness is created
in the interaction between things.
- Yes, I agree.
- So the joy is in the creation of consciousness.
- I see.
- I really like the idea that it doesn't just have to be
two humans creating consciousness together.
It could be humans and other entities.
We talked offline about dogs and other pets and so on.
There's a magic, I mean, I've been calling it love.
There's beauty of the human experience that's created
and it just feels like fascinating that you could do that
with a robotic system.
There's something really powerful at least to me
about engineering systems that allow you to create
some of the magic of the human experience
'cause then you get to understand what it takes,
at least get inklings of what it takes
to create consciousness.
And I don't get this, you know,
philosophers get really upset about this idea
that sort of the illusion of consciousness is consciousness.
But I really like the idea of engineering systems
that fool you into thinking they're conscious.
- Right.
- Because that's sufficient to create
the magical experience.
- Right, because it's the interaction, yeah.
- It's the interaction, yeah.
And this is the Russian hat I wear,
which is like, I think there's an ocean
of loneliness in the world.
I think we're deeply lonely.
We're not even allowing ourselves to acknowledge that.
And I kind of think that's what love is
between romantic love and friendship
is two people kind of getting a little bit
alleviating for a brief moment.
- That loneliness.
- That loneliness, but we're not,
it's not the full aspect of that loneliness.
Like we're desperately alone,
we're desperately afraid of non-existing.
- Right.
- I have that kind of sense,
and I just wanna explore that ocean of loneliness more.
- Right.
- In engineering, like create a submarine
that goes into the depth of that loneliness.
So creating systems that can truly hear you.
- Right.
- And truly listen.
- Make the universe a less lonely place.
- Exactly.
Let me ask you about the meaning.
You've brought up why.
- Yeah.
- The physics of why.
What do you think is the meaning of our particular planets,
set of existences, and the universe in general?
The meaning of life.
- Yes, someone once told me as a physicist,
I'm not allowed to ask why questions,
but I don't believe that.
I think what we are is the creative process
in the universe, I think.
And for me, that's the meaning.
The ability to-- - Create.
- Yeah, to create more possibilities
and more things to exist.
- What is, Odessiyevsky has the saying,
"Beauty will save the world."
Is there a connection between creation and beauty?
- I think so.
- So is that like, is beauty a correlate of creation?
- It might be, I don't know.
I mean, why is it, you know,
a lot of people have asked these kind of questions,
but like, why is it we have such an emotional response
to intellectual activity or creativity?
And that seems kind of a deep question to me.
Like, it seems very intrinsic to what we are.
So I do have an interest in the questions I ask
because I think they're beautiful,
and I think the universe is beautiful,
and I'm just so deeply fascinated
by the fact that I exist at all.
And so maybe it's that, you know,
that intrinsic feeling of beauty that's in part driving,
you know, the physics of creating more things,
so they could be deeply related in that way.
- Well, I don't think there's a better way to end it.
I think this conversation was beautiful.
Thank you so much for wasting
all your valuable time with me today.
I really, really appreciate it, Sarah.
This is an honor.
I hope we get the chance to talk again.
I hope, like I mentioned to you offline,
we get a chance to talk with Lee.
You guys have a beautiful, like, intellectual chemistry
that's fascinating to listen to,
so I'm a huge fan of both of you,
and I can't wait to see what you do next.
- Thanks so much.
Great to be here.
Bye. - Bye.
- Thanks for listening to this conversation
with Sarah Walker.
A thank you to Athletic Greens, NetSuite,
Blinkist, and Magic Spoon.
Check them out in the description to support this podcast.
And now let me leave you with some words from Robert Frost,
one of my favorite poets.
In three words,
I can sum up everything I've learned about life.
It goes on.
Thank you for listening.
I hope to see you next time.
(upbeat music)
(upbeat music)