back to indexLeonard Susskind: Quantum Mechanics, String Theory and Black Holes | Lex Fridman Podcast #41
link |
The following is a conversation with Leonard Suskind.
link |
He's a professor of theoretical physics at Stanford University
link |
and founding director of Stanford Institute of Theoretical Physics.
link |
He is widely regarded as one of the fathers of string theory and in general
link |
is one of the greatest physicists of our time both as a researcher
link |
and an educator. This is the Artificial Intelligence Podcast.
link |
Perhaps you noticed that the people I've been speaking with are not just
link |
computer scientists but philosophers, mathematicians, writers, psychologists,
link |
physicists, and soon other disciplines. To me,
link |
AI is much bigger than deep learning, bigger than computing.
link |
It is our civilization's journey into understanding the human mind
link |
and creating echoes of it in the machine. If you enjoy the podcast, subscribe on
link |
YouTube, give it five stars on iTunes, support on Patreon,
link |
or simply connect with me on Twitter at Lex Freedman, spelled F R I D M A M.
link |
And now, here's my conversation with Leonard Suskind.
link |
You worked and were friends with Richard Feynman. How has he influenced you
link |
and changed you as a physicist and thinker?
link |
What I saw, I think what I saw was somebody who could do physics in this
link |
deeply intuitive way. His style was almost to close his
link |
eyes and visualize the phenomena that he was thinking about.
link |
And through visualization,
link |
outflank the mathematical, the highly mathematical and
link |
very, very sophisticated technical arguments that people would use.
link |
I think that was also natural to me, but I saw somebody who was actually
link |
successful at it, who could do physics in a way
link |
that I regarded as simpler, more direct,
link |
more intuitive. And while I don't think he changed
link |
my way of thinking, I do think he validated it.
link |
He made me look at it and say, yeah, that's something you can do
link |
and get away with. Practically, you can get away with it.
link |
So, do you find yourself, whether you're thinking about quantum mechanics
link |
or black holes or string theory using intuition
link |
as a first step or step throughout using visualization?
link |
Yeah, very much so. Very much so. I tend not to think about the equations. I tend
link |
not to think about the symbols. I tend to try to visualize the
link |
phenomena themselves. And then when I get an insight
link |
that I think is valid, I might try to convert it to mathematics. But I'm not a
link |
I'm not a natural mathematician or I'm good enough at it.
link |
I'm good enough at it, but I'm not a great mathematician.
link |
So, for me, the way of thinking about physics is first intuitive, first
link |
visualization, scribble a few equations maybe, but then try to convert it to
link |
mathematics. Experiences that other people are
link |
better at converting it to mathematics than I am.
link |
And yet, you've worked very counterintuitive ideas.
link |
So, how do you visualize something counterintuitive?
link |
How do you dare rewiring your brain in new ways?
link |
Yeah, quantum mechanics is not intuitive. Very little of modern physics is
link |
intuitive. What does intuitive mean?
link |
It means the ability to think about it with basic classical physics, the
link |
physics that we evolved with throwing stones,
link |
splashing water, whatever it happens to be. Quantum physics, general relativity,
link |
quantum field theory are deeply unintuitive in that way.
link |
But, you know, after time and getting familiar with these things, you
link |
develop new intuitions. I always said you rewire.
link |
And it's to the point where me and many of my friends,
link |
I and many of my friends, can think more easily quantum
link |
mechanically than we can classically. We've gotten so used to it.
link |
I mean, yes, our neural wiring in our brain is such that we understand
link |
rocks and stones and water and so on. We're sort of evolved for that.
link |
Evolved for it. Do you think it's possible to create a wiring
link |
of neuron like state devices that more naturally understand
link |
quantum mechanics, understand wave function, understand these weird things?
link |
Well, I'm not sure. I think many of us have evolved
link |
the ability to think quantum mechanically to some extent.
link |
But that doesn't mean you can think like an electron.
link |
That doesn't mean another example. Forget for a minute, quantum mechanics.
link |
Just visualizing four dimensional space or five dimensional space or six
link |
dimensional space, I think we're fundamentally wired
link |
to visualize three dimensions. I can't even
link |
visualize two dimensions or one dimension without thinking about it as
link |
embedded in three dimensions. If I want to visualize a line,
link |
I think of the line as being a line in three dimensions.
link |
Or I think of the line as being a line on a piece of paper with a piece of paper
link |
being in three dimensions. I never seem to be able to
link |
in some abstract and pure way visualize in my head
link |
the one dimension, the two dimension, the four dimension, the five dimensions,
link |
and I don't think that's ever going to happen. The reason is I think
link |
neural wiring is just set up for that.
link |
On the other hand, we do learn ways to think about
link |
five, six, seven dimensions. We learn ways, we learn
link |
mathematical ways, and we learn ways to visualize them, but they're different.
link |
And so yeah, I think we do rewire ourselves, whether we can ever
link |
completely rewire ourselves to be completely comfortable with these
link |
concepts, I doubt. So that it's completely natural.
link |
Difficult to where it's completely natural. So I'm sure there's
link |
somewhat, you could argue, creatures that live in a two dimensional
link |
space. Yeah, maybe there are. And while it's
link |
romanticizing the notion, of course, we're all living as far as we know in
link |
three dimensional space, but how do those creatures imagine
link |
3D space? Well, probably the way we imagine 4D by
link |
using some mathematics and some equations and some
link |
some tricks. Okay, so jumping back to Feynman just
link |
for a second, he had a little bit of an ego.
link |
Yes. Do you think ego is powerful or dangerous in science?
link |
I think both. Both. I think you have to have both arrogance
link |
and humility. You have to have the arrogance to say
link |
I can do this. Nature is difficult. Nature is very, very hard.
link |
I'm smart enough. I can do it. I can win the battle with nature.
link |
On the other hand, I think you also have to have the humility
link |
to know that you're very likely to be wrong on any given occasion.
link |
Everything you're thinking could suddenly change.
link |
Young people can come along and say things you won't understand and you'll be
link |
lost and flabbergasted. So I think it's a combination of both.
link |
You better recognize that you're very limited
link |
and you better be able to say to yourself I'm not so limited
link |
that I can't win this battle with nature. It takes a special kind of
link |
person who can manage both of those, I would say.
link |
And I would say there's echoes of that in your own work.
link |
A little bit of ego, a little bit of outside of the box, humble thinking.
link |
I hope so. So was there a time
link |
where you felt you looked at yourself and asked
link |
am I completely wrong about this? Oh yeah, about the whole thing or about
link |
specific things. The whole thing? What do you mean?
link |
Wait, which whole thing? Me and me and my ability to do this thing.
link |
Oh, those kinds of doubts. First of all, did you have those kinds of doubts?
link |
No, I had different kind of doubts. I came from a very working class background
link |
and I was uncomfortable in academia for, well, for a long time.
link |
But there weren't doubts about my ability or my,
link |
they were just the discomfort and being in an environment that
link |
my family hadn't participated in. I knew nothing about as a young person. I
link |
didn't learn that there was such a thing called physics until I was almost 20
link |
years old. So I did have certain kind of
link |
doubts, but not about my ability. I don't think I was too
link |
worried about whether I would succeed or not.
link |
I never felt this insecurity, am I ever going to get a job?
link |
That had never occurred to me that I wouldn't.
link |
Maybe you could speak a little bit to this sense of what is academia,
link |
because I too feel a bit uncomfortable in it.
link |
There's something I can't put quite into words, what you have
link |
that's not, doesn't, if we call it music,
link |
you play a different kind of music than a lot of academia.
link |
How have you joined this orchestra? How do you think about it?
link |
I don't know that I thought about it as much as I just
link |
felt it. You know, thinking is one thing, feeling is another thing.
link |
I felt like an outsider until a certain age when I suddenly found myself the
link |
ultimate insider in academic physics.
link |
That was a sharp transition and I wasn't a young man. I was
link |
probably 50 years old. You were never quite, it was a face
link |
transition, you were never quite in the middle.
link |
Yeah, that's right, I wasn't. I always felt a little bit of an outsider
link |
in the beginning, a lot an outsider.
link |
My way of thinking was different, my approach to mathematics was different,
link |
but also my social background that I came from was different.
link |
Now these days, half the young people I meet, they're parents or professors.
link |
Right. That was not my case.
link |
So yeah, but then all of a sudden at some point I found myself at very much the
link |
center of, maybe not the only one at the center, but certainly one of the people
link |
in the center of a certain kind of physics and all that went away.
link |
I mean it went away in a flash. So maybe a little bit with Feynman,
link |
but in general, how do you develop ideas? Do you work
link |
through ideas alone? Do you brainstorm with others?
link |
Oh, both, both, very definitely both. The younger time I spent more time with
link |
myself. Now because I'm at
link |
Stanford, because I'm, because I have a lot of
link |
ex students and people who are interested in the same thing I am,
link |
I spend a good deal of time almost on a daily basis
link |
interacting, brainstorming, as you said. It's a very important part.
link |
I spend less time probably completely self focused than
link |
with a piece of paper and just sitting there staring at it.
link |
What are your hopes for quantum computers?
link |
So machines that are based on, that have some elements of
link |
leveraged quantum mechanical ideas. Yeah, it's not just leveraging quantum
link |
mechanical ideas. You can simulate quantum systems on a classical computer.
link |
Simulate them means solve this Schrodinger equation for them
link |
or solve the equations of quantum mechanics on
link |
a computer, on a classical computer. But the classical computer is not doing,
link |
is not a quantum mechanical system itself. Of course it is. Everything's made of
link |
quantum mechanics, but it's not functioning. It's not functioning as
link |
a quantum system. It's just solving equations.
link |
The quantum computer is truly a quantum system which is
link |
actually doing the things that you're programming it to do.
link |
You want to program a quantum field theory?
link |
If you do it in classical physics, that program is not actually functioning in
link |
the computer as a quantum field theory. It's just solving some equations.
link |
Physically it's not doing the things that the quantum system would do.
link |
The quantum computer is really a quantum mechanical system which is
link |
actually carrying out the quantum operations. You can measure it at the end.
link |
It intrinsically satisfies the uncertainty principle.
link |
It is limited in the same way that quantum systems are limited
link |
by uncertainty and so forth. And it really is a quantum system. That means
link |
that what you're doing when you program something for a quantum
link |
system is you're actually building a real version of the system.
link |
The limits of a classical computer. Classical computers are enormously limited
link |
when it comes to the quantum systems. They're enormously
link |
limited because you've probably heard this before.
link |
But in order to store the amount of information that's in the quantum state
link |
of 400 spins, that's not very many. 400 can put in my pocket.
link |
400 pennies in my pocket.
link |
To be able to simulate the quantum state of 400 elementary quantum
link |
systems, qubits we call them, to do that would take more information than can
link |
possibly be stored in the entire universe if it were packed
link |
so tightly that you couldn't pack any more in.
link |
400 qubits. On the other hand, if your quantum
link |
computer is composed of 400 qubits, it can do everything 400 qubits can do.
link |
What kind of space, if you just intuitively think about the space of
link |
algorithms that that unlocks for us. So there's a whole complexity theory
link |
around classical computers measuring the running time of things
link |
and p, so on. What kind of algorithms just intuitively do you think it unlocks for us?
link |
Okay, so we know that there are a handful of algorithms that can seriously
link |
be classical computers and which can have exponentially
link |
more power. This is a mathematical statement. Nobody's exhibited this in
link |
the laboratory. That's a mathematical statement. We know
link |
that's true, but it also seems more and more that the
link |
number of such things is very limited. Only very, very special
link |
problems exhibit that much advantage for a quantum computer,
link |
of standard problems. To my mind, as far as I can tell, the great power of
link |
quantum computers will actually be the simulate quantum systems.
link |
If you're interested in a certain quantum system
link |
and it's too hard to simulate classically, you simply build a version
link |
of the same system. You build a version of it, you build a model of it that's
link |
actually functioning as the system, you run it, and then you do the same thing
link |
you would do the quantum system, you make measurements on it,
link |
quantum measurements on it. The advantage is you can run it much slower
link |
because say, why bother? Why not just use the real system?
link |
Why not just do experiments on the real system? Well, real systems are kind of
link |
limited. You can't change them. You can't manipulate them.
link |
You can't slow them down so that you can poke into them.
link |
You can't modify them in arbitrary kinds of ways to see what would happen
link |
if I change the system a little bit. I think that quantum computers
link |
will be extremely valuable in understanding quantum
link |
systems. At the lowest level, the fundamental
link |
laws. They're actually satisfying the same laws
link |
as the systems that they're simulating. That's right.
link |
Okay, so on the one hand you have things like factoring.
link |
Factoring is the great thing of quantum computers, factoring large
link |
numbers. That doesn't seem that much to do with
link |
quantum mechanics. It seems to be almost a fluke
link |
that a quantum computer can solve the factoring problem
link |
in a short time. Those problems seem to be extremely special,
link |
rare, and it's not clear to me that there's going to be a lot of them.
link |
On the other hand, there are a lot of quantum systems. There's chemistry,
link |
there's solid state physics, there's material science,
link |
there's quantum gravity, there's all kinds of quantum field theory.
link |
Some of these are actually turning out to be applied sciences as well as very
link |
fundamental sciences. We probably will run out of the
link |
ability to solve equations for these things.
link |
Solve equations by the standard methods of pencil and paper.
link |
Solve the equations by the method of classical computers.
link |
And so what we'll do is we'll build versions of these systems,
link |
run them, and run them under controlled circumstances where we can change them,
link |
manipulate them, make measurements on them, and find out all the things we want
link |
to know. So in finding out the things we want to know
link |
about very small systems, is there something we can also find out
link |
about the macro level, about something about the function,
link |
forgive me, of our brain, biological systems, the stuff that's about one meter
link |
in size versus much, much smaller? Well, what all the
link |
excitement is about among the people that I interact with is
link |
understanding black holes. Black holes. Black holes are big things.
link |
They are many, many degrees of freedom. There is another kind of quantum system
link |
that is big. It's a large quantum computer. And one of the things we've
link |
learned is that the physics of large quantum computers is in some way
link |
similar to the physics of large quantum black holes.
link |
And we're using that relationship. Now you asked, you didn't ask about quantum
link |
computers as systems, you didn't ask about black holes, you asked about brains.
link |
Yeah, about stuff that's in the middle of the two. It's different.
link |
So black holes are, there's something fundamental
link |
about black holes that feels to be very different than a brain. Yes.
link |
And they also function in a very quantum mechanical way. Right.
link |
Okay. It is, first of all, unclear to me, but of course it's unclear to me. I'm not
link |
I'm not a neuroscientist. I have, I don't even have very many friends who are
link |
neuroscientists. I would like to have more friends who are
link |
neuroscientists. I just don't run into them very often.
link |
Among the few neuroscientists I've ever talked about about this,
link |
they are pretty convinced that the brain
link |
functions classically, that it is not intrinsically a quantum mechanical
link |
system, or it doesn't make use of the of the
link |
special features entanglement, coherent superposition.
link |
Are they right? I don't know. I sort of hope they're wrong,
link |
just because I like the romantic idea that the brain is a quantum
link |
system. Yeah. But I think, I think probably not.
link |
The other thing, big systems can be composed of lots of little systems.
link |
Okay. Materials, the materials that we work with and so forth
link |
are, can be large systems, a large piece of material,
link |
but they're big and they're made out of quantum systems. Now one of the things
link |
that's been happening over the last, a good number of years
link |
is we're discovering materials and quantum systems
link |
which function much more quantum mechanically
link |
than we imagine. Topological insulators, this kind of thing, that kind of thing.
link |
Those are macroscopic systems, but they're just superconductors,
link |
superconductors. I have a lot of quantum mechanics in them.
link |
You can have a large chunk of superconductor, so it's a big piece of
link |
material. On the other hand, it's functioning and
link |
its properties depend very, very strongly on quantum mechanics
link |
and to analyze them, you need the tools of quantum mechanics.
link |
If we can go on to black holes and looking at the universe as a
link |
information processing system, as a computer, as a giant computer,
link |
what's the power of thinking of the universe as an information processing
link |
system, but what is perhaps its use besides the
link |
mathematical use of discussing black holes and
link |
your famous debates and ideas around that to
link |
human beings or life in general as information processing systems.
link |
But all systems are information processing systems.
link |
You poke them, they change a little bit, they evolve.
link |
All systems are information processing systems. So there's no extra magic
link |
to us humans. It certainly feels, consciousness and
link |
intelligence feels like magic. It sure does. Where does it emerge from
link |
if we look at information processing?
link |
What are the emergent phenomena that come from viewing the world
link |
as an information processing system? Here is what I think.
link |
My thoughts are not worth much in this. If you ask me about physics, my thoughts
link |
may be worth something. Yes. If you ask me about this, I'm not
link |
sure my thoughts are worth anything, but as I said earlier,
link |
I think when we do introspection, when we imagine doing introspection and try to
link |
figure out what it is when we do and we're thinking,
link |
I think we get it wrong. I'm pretty sure we get it wrong.
link |
Everything I've heard about the way the brain functions is so counterintuitive.
link |
For example, you have neurons which detect vertical lines.
link |
You have different neurons which detect lines at 45 degrees. You have different
link |
neurons. I never imagined that there were whole
link |
circuits which were devoted to vertical lines in my brain.
link |
It doesn't seem to me the way my brain works. My brain seems to work if I put my
link |
finger up vertically, or if I put it horizontally, or if I put it this way or
link |
that way. It seems to me it's the same circuits.
link |
It's not the way it works. The way the brain is compartmentalized
link |
seems to be very, very different than what I would have imagined if I were just
link |
doing psychological introspection about how things work.
link |
My conclusion is that we won't get it right that way.
link |
How will we get it right? I think maybe computer scientists will get it
link |
right eventually. I don't think there are any ways near it. I don't even think
link |
they're thinking about it. But by computer, eventually we will
link |
build machines perhaps which are complicated enough
link |
and partly engineered, partly evolved, maybe evolved by machine learning and so
link |
forth. This machine learning is very interesting.
link |
By machine learning we will evolve systems and we may start
link |
to discover mechanisms that have implications
link |
for how we think and for what this consciousness thing is all about.
link |
We'll be able to do experiments on them and perhaps answer questions
link |
that we can't possibly answer by introspection.
link |
That's a really interesting point. In many cases,
link |
if you look even at string theory, when you first think about a system it seems
link |
really complicated, like the human brain. Through some basic reasoning
link |
and trying to discover fundamental low level
link |
behavior of the system, you find out that it's actually much simpler.
link |
Is that generally the process and do you have that also hope
link |
for biological systems as well, for all the kinds of stuff we're studying
link |
at the human level? Of course, physics always begins by trying to find the
link |
simplest version of something and analyze it. There are lots of
link |
examples where physics has taken very complicated
link |
systems, analyzed them and found simplicity in them for sure.
link |
I said superconductors before. It's an obvious one. A superconductor seems like
link |
a monstrously complicated thing with all sorts of crazy
link |
electrical properties, magnetic properties and so forth.
link |
When it finally is boiled down to its simplest elements,
link |
it's a very simple quantum mechanical phenomenon
link |
called spontaneous symmetry breaking which we, in other contexts,
link |
we learned about and we're very familiar with. So yeah, I mean,
link |
yes, we do take complicated things, make them simple,
link |
but what we don't want to do is take things which are intrinsically
link |
complicated and fool ourselves into thinking that we can make them simple.
link |
I don't know who said this, but we don't want to make them simpler than they
link |
really are. Is the brain a thing which ultimately
link |
functions by some simple rules or is it just complicated?
link |
In terms of artificial intelligence, nobody really knows
link |
what are the limits of our current approaches. You mentioned machine
link |
learning. How do we create human level intelligence? It seems that there's a
link |
lot of very smart physicists who perhaps oversimplify the nature of
link |
intelligence and think of it as information processing
link |
and therefore there doesn't seem to be any theoretical reason
link |
why we can't artificially create human level or superhuman level
link |
intelligence. In fact, the reasoning goes if you create human level intelligence,
link |
the same approach you just used to create human level intelligence
link |
should allow you to create superhuman level intelligence very easily,
link |
exponentially. So what do you think that way of thinking
link |
that comes from physicist is all about? I wish I knew, but there's a particular
link |
reason why I wish I knew. I have a second job.
link |
I consult for Google, not for Google, for Google X.
link |
I am the senior academic advisor to a group of machine learning physicists.
link |
Now that sounds crazy because I know nothing about the subject.
link |
I know very little about the subject. On the other hand, I'm good at giving
link |
advice, so I give them advice on things. Anyway,
link |
I see these young physicists who are approaching the machine learning problem.
link |
There is a real machine learning problem, namely why does it
link |
work as well as it does? Nobody really seems to understand
link |
why it is capable of doing the kind of generalizations that it does
link |
and so forth. There are three groups of people who
link |
have thought about this. There are the engineers.
link |
The engineers are incredibly smart, but they tend not to think as hard about
link |
why the thing is working as much as they do how to use it.
link |
Obviously, they provided a lot of data, and it is they who demonstrated that
link |
machine learning can work much better than you had any right to expect.
link |
The machine learning systems are systems that the system is not too
link |
different than the kind of systems that physicists study.
link |
There's not all that much difference in the structure of the mathematics.
link |
Physically, yes, but in the structure of the mathematics
link |
between a tensor network designed to describe a quantum system, on the one
link |
hand, and the kind of networks that are used in
link |
machine learning. So, there are more and more, I think,
link |
young physicists are being drawn to this field of machine learning.
link |
Some very, very good ones. I work with a number of very good ones.
link |
Not on machine learning, but on having lunch.
link |
On having lunch, right. And I can tell you, they are super smart.
link |
They don't seem to be so arrogant about their physics backgrounds that they
link |
think they can do things that nobody else can do.
link |
But the physics way of thinking, I think, will add
link |
great value to, will bring value to the machine learning.
link |
I believe it will. And I think it already has.
link |
At what time scale do you think predicting the future becomes useless?
link |
And your long experience and being surprised in your discoveries?
link |
Sometimes a day, sometimes 20 years. There are things which I thought
link |
we were very far from understanding, which practically in a snap of the fingers
link |
or a blink of the eye suddenly became understood,
link |
completely surprising to me.
link |
There are other things which I looked at and I said,
link |
we're not going to understand these things for 500 years,
link |
in particular quantum gravity. The scale for that was 20 years, 25 years.
link |
And we understand a lot, and we don't understand it completely now by any
link |
means. But I thought it was 500 years to make any
link |
progress. It turned out to be very, very far from
link |
that. It turned out to be more like 20 or 25 years
link |
from the time when I thought it was 500 years.
link |
So if we may, can we jump around quantum gravity, some basic ideas in physics?
link |
What is the dream of string theory, mathematically?
link |
What is the hope? Where does it come from? What problem is it trying to solve?
link |
I don't think the dream of string theory is any different than the dream of
link |
fundamental theoretical physics altogether. Understanding a unified
link |
theory of everything. I don't like thinking of string theory
link |
as a subject unto itself with people called string theorists who are the
link |
practitioners of this thing called string theory.
link |
I much prefer to think of them as theoretical physicists
link |
trying to answer deep fundamental questions about nature,
link |
in particular gravity, in particular gravity and its connection with quantum
link |
mechanics, and who at the present time find
link |
string theory a useful tool rather than saying there's a subject
link |
called string theorist. I don't like being referred to as a string
link |
theorist. Yes, but as a tool, is it useful to think
link |
about our nature in multiple dimensions, the strings
link |
vibrating? I believe it is useful. I'll tell you
link |
what the main use of it has been up till now. Well, it has had a number of main
link |
uses. Originally string theory was invented,
link |
then I know there I was there. I was right at the spot where it was being
link |
invented, literally, and it was being invented to
link |
understand hadrons. Hadrons are subnuclear particles,
link |
protons, neutrons, mesons, and at that time, the late 60s,
link |
early 70s, it was clear from experiment that these particles
link |
called hadrons could vibrate, could rotate,
link |
could do all the things that a little closed string can do,
link |
and it was and is a valid and correct theory of these
link |
hadrons. It's been experimentally tested, and that is a done deal.
link |
It had a second life as a theory of gravity, the same basic mathematics
link |
except on a very, very much smaller distance scale.
link |
The objects of gravitation are 19 orders of magnitude smaller than
link |
a proton, but the same mathematics turned up.
link |
The same mathematics turned up. What has been its value?
link |
Its value is that it's mathematically rigorous in many ways
link |
and enabled us to find mathematical structures which have
link |
both quantum mechanics and gravity with rigor.
link |
We can test out ideas. We can test out ideas. We can't test them in the
link |
laboratory. They had 19 orders of magnitude too
link |
small, the things that we're interested in, but we can test them out
link |
mathematically and analyze their internal consistency.
link |
By now, 40 years ago, 35 years ago, so forth, people very, very much
link |
questioned the consistency between gravity and quantum mechanics.
link |
Stephen Hawking was very famous for it, rightly so.
link |
Now nobody questions that consistency anymore. They don't because we have
link |
mathematically precise string theories which contain both
link |
gravity and quantum mechanics in a consistent way.
link |
So it's provided that certainty that quantum mechanics and gravity can
link |
coexist. That's not a small thing. It's a very big thing.
link |
It's a huge thing. Einstein will be proud.
link |
Einstein, he might be appalled. I don't know. He didn't like quantum mechanics very much,
link |
but he would certainly be struck by it. Yeah.
link |
I think that may be at this time its biggest contribution to physics in
link |
illustrating almost definitively that quantum mechanics and gravity are very
link |
closely related and not inconsistent with each other.
link |
Is there a possibility of something deeper, more profound,
link |
that still is consistent with string theory but is deeper
link |
that is to be found? Well, you could ask the same thing about quantum mechanics.
link |
Is there something? Exactly.
link |
Yeah. I think string theory is just an example of a quantum mechanical system
link |
that contains both gravitation and quantum mechanics.
link |
So is there something underlying quantum mechanics?
link |
Perhaps something deterministic.
link |
Perhaps something deterministic. My friend,
link |
Ferrad Etouft, whose name you may know, he's a very famous physicist.
link |
Dutch, not as famous as he should be, but...
link |
Hard to spell his name. It's hard to say his name.
link |
No, it's easy to spell his name. Apostrophe. He's the only person I know whose
link |
name begins with a metastrophe. And he's one of my heroes in physics.
link |
He's a little younger than me, but he's nevertheless one of my heroes.
link |
Etouft believes that there is some sub structure to the world
link |
which is classical in character, deterministic in character,
link |
which somehow by some mechanism that he has a hard time
link |
spelling out emerges as quantum mechanics.
link |
I don't. The wave function is somehow emergent.
link |
The wave function and not just the wave function, but the whole
link |
mech and the whole thing that goes with quantum mechanics, uncertainty,
link |
entanglement, all these things are emergent.
link |
So you think quantum mechanics is the bottom of the well?
link |
Is the... Here I think is where you have to be humble.
link |
Here's where humility comes. I don't think anybody should say anything is
link |
the bottom of the well at this time. I think we...
link |
I think we can reasonably say...
link |
I can reasonably say when I look into the well,
link |
I can't see past quantum mechanics. I don't see any reason for it to be
link |
anything beyond quantum mechanics. I think Etouft has asked very
link |
interesting and deep questions. I don't like his answers.
link |
Well, again, let me ask, if we look at the deepest nature of reality
link |
with whether it's deterministic or when observed as probabilistic,
link |
what does that mean for our human level of ideas of free will?
link |
Is there any connection whatsoever from this perception, perhaps illusion, of
link |
free will that we have and the fundamental nature of reality?
link |
The only thing I can say is I am puzzled by that as much as you are.
link |
The illusion of it. The illusion of consciousness, the illusion of free will,
link |
the illusion of self. Does that connect to...
link |
How can a physical system do that? And I am as puzzled as anybody.
link |
There's echoes of it in the observer effect.
link |
Yeah. So do you understand what it means to be an observer?
link |
I understand it at a technical level. An observer is a system with enough
link |
degrees of freedom that it can record information and which can become
link |
entangled with the thing that it's measuring. Entanglement is the key.
link |
When a system which we call an apparatus or an observer,
link |
same thing, interacts with the system that it's observing,
link |
it doesn't just look at it, it becomes physically entangled with it.
link |
And it's that entanglement which we call an observation or a measurement.
link |
Now, does that satisfy me personally as an observer?
link |
Yes or no. I find it very satisfying that we have a mathematical representation
link |
of what it means to observe a system. You are observing stuff right now,
link |
the conscious level. Right. Do you think there's echoes
link |
of that kind of entanglement in our macro scale?
link |
Yes, absolutely, for sure. We're entangled with it,
link |
quantum mechanically entangled with everything in this room.
link |
If we weren't, then it would just, well, we wouldn't be observing it.
link |
But on the other hand, you can ask, am I really comfortable with it?
link |
And I'm uncomfortable with it in the same way that I can never get comfortable
link |
with five dimensions. My brain isn't wired for it.
link |
Are you comfortable with four dimensions? A little bit more,
link |
because I can always imagine the fourth dimension is time.
link |
So the arrow of time, are you comfortable with that arrow?
link |
Do you think time is an emergent phenomena or is it fundamental to nature?
link |
That is a big question in physics right now. All the physics that we do,
link |
or at least at the people that I am comfortable talking to,
link |
my friends, my friends. No, we all ask the same question that you just asked.
link |
Space, we have a pretty good idea, is emergent and it emerges out of
link |
entanglement and other things. Time always seems to be built into
link |
our equations as just what Newton pretty much would have thought.
link |
Newton modified a little bit by Einstein, would have called time.
link |
And mostly in our equations, it is not emergent.
link |
Time in physics is completely symmetric forward and back.
link |
Symmetric. So you don't really need to think about
link |
the arrow of time for most physical phenomena.
link |
For most microscopic phenomena, no. It's only when the phenomena involves
link |
systems which are big enough for thermodynamics to become important,
link |
for entropy to become important. For a small system,
link |
entropy is not a good concept.
link |
Entropy is something which emerges out of large numbers.
link |
It's a probabilistic idea, it's a statistical idea,
link |
and it's a thermodynamic idea. Thermodynamics requires lots and lots and
link |
lots of little substructures. So it's not until you emerge
link |
at the thermodynamic level that there's an arrow of time,
link |
do we understand it? Yeah, I think we understand better than most people
link |
think they are. Most people say they think we understand it.
link |
Yeah, I think we understand it. It's a statistical idea.
link |
You mean like second law, thermodynamics, entropy and so on?
link |
Yeah, you take a pack of cards and you fling it in the air and you look what
link |
happens to it. It gets random. It doesn't go from
link |
random to simple. It goes from simple to random.
link |
But do you think it ever breaks down? What I think
link |
you can do is in a laboratory setting, you can take a system which is somewhere
link |
intermediate between being small and being large
link |
and make it go backward. A thing which looks like it
link |
only wants to go forward because of statistical mechanical reasons, because of
link |
the second law, you can very very carefully manipulate it
link |
to make it run backward. I don't think you can take an egg
link |
Humpty Dumpty who fell on the floor and reverse that,
link |
but you can in a very controlled situation, you can take systems
link |
which appear to be evolving statistically toward randomness,
link |
stop them, reverse them and make them go back.
link |
What's the intuition behind that? How do we do that? How do we reverse it?
link |
You're saying a closed system. Yeah, pretty much closed system.
link |
Yes. Did you just say that time travel is possible?
link |
No, I didn't say time travel is possible. I said you can make a system go
link |
backward. In time. You can make it go back. You can
link |
make it reverse its steps. You can make it reverse its trajectory.
link |
Yeah. How do we do it? What's the intuition there? Is it just a
link |
fluke thing that we can do at a small scale in
link |
the lab that doesn't have... What I'm saying is you can do it on a little
link |
bit better than a small scale. You can certainly do it
link |
with a simple small system. Small systems don't have any sense of the
link |
arrow of time atoms. Atoms are
link |
no sense of an arrow of time. They're completely reversible.
link |
It's only when you have... The second law of thermodynamics is the
link |
law of large numbers. So you can break the law because it's not...
link |
You can break it, but it's hard. It requires great
link |
care. The bigger the system is, the more care and more
link |
the harder it is. You have to overcome what's called chaos.
link |
And that's hard. And it requires more and more precision. For 10 particles, you
link |
might be able to do it with some effort.
link |
For 100 particles, it's really hard. For a thousand or a million particles,
link |
forget it. But not for any fundamental reason, just
link |
because it's technologically too hard to make the system go
link |
So no time travel for engineering reasons?
link |
No. What is time travel? Time travel to the future?
link |
That's easy. You just close your eyes, go to sleep, and you wake up in the
link |
future. Good nap gets you there.
link |
But in reversing the second law of thermodynamics,
link |
going backward in time for anything that's human scale is a very difficult
link |
engineering effort. I wouldn't call that time travel because
link |
it gets too mixed up with what science fiction calls time travel.
link |
This is just the ability to reverse a system. You take the system
link |
and you reverse the direction of motion of every molecule in it.
link |
You can do it with one molecule. If you find a particle moving in a
link |
certain direction, let's not say a particle, a baseball,
link |
you stop it dead and then you simply reverse its motion
link |
in principle. That's not too hard and it'll go back along its
link |
trajectory in the backward direction. Just running the program backwards?
link |
Running the program backward. If you have two baseballs colliding,
link |
well you can do it but you have to be very very careful to get it just
link |
right. If you have 10 baseballs, really really
link |
better yet, 10 billiard balls on an idealized frictionless billiard
link |
table. You start the balls all in a triangle
link |
and you whack them. Depending on the game you're playing, you either whack them
link |
or you're really careful but you whack them and they go flying
link |
off in all possible directions. Try to reverse that.
link |
Imagine trying to take every billiard ball stopping it dead at some
link |
point and reversing its motion so it was going in the opposite direction.
link |
If you did that with tremendous care, it would
link |
reassemble itself back into the triangle. Okay,
link |
that is a fact and you can probably do it with two billiard balls, maybe with
link |
three billiard balls if you're really lucky. But what happens is as the system
link |
gets more and more complicated, you have to be more and more precise
link |
not to make the tiniest error because the tiniest errors will get magnified
link |
and you'll simply not be able to do the reversal.
link |
So yeah, but I wouldn't call that time travel.
link |
Yeah, that's something else. But if you think of it,
link |
it just made me think, if you think the unrolling
link |
of state that's happening as a program, if we look at the world,
link |
silly idea of looking at the world as a simulation,
link |
as a computer, but it's not a computer, it's just a single program.
link |
A question arises that might be useful, how hard is it
link |
to have a computer that runs the universe? Okay, so
link |
there are mathematical universes that we know about, one of them is called
link |
anti decider space where we and its quantum mechanics.
link |
I think we could simulate it in a quantum computer.
link |
Classical computer, all you can do is solve its equations. You can't make it
link |
work like the real system. If we could build a quantum
link |
computer, a big enough one, a robust enough one,
link |
we could probably simulate a universe,
link |
a small version of an anti decider universe. Anti decider is a kind of a
link |
cosmology. So I think we know how to do that.
link |
The trouble is the universe that we live in is not
link |
the anti decider geometry, it's the decider geometry,
link |
and we don't really understand its quantum mechanics at all.
link |
So at the present time, I would say we wouldn't have the vaguest idea how to
link |
simulate a universe similar to our own.
link |
Could we build in the laboratory a small version,
link |
a quantum mechanical version, the collection of quantum computers entangled
link |
and coupled together, which would reproduce the phenomena that go on
link |
in the universe, even on a small scale. Yes, if it were anti decider space,
link |
no, if it's the decider space. Can you slightly describe the decider space and
link |
anti decider space? Yeah.
link |
What are the geometric properties of? They differ by the sign of a single
link |
constant called the cosmological constant. One of them
link |
is negatively curved, the other is positively curved.
link |
Anti decider space, which is the negatively curved one,
link |
you can think of as an isolated system in a box with reflecting walls.
link |
You could think of it as a system of quantum mechanical system,
link |
isolated in an isolated environment. The decider space is the one we really
link |
live in, and that's the one that's exponentially expanding.
link |
Exponential expansion, dark energy, whatever we want to call it,
link |
and we don't understand that mathematically.
link |
Not everybody would agree with me, but I don't understand.
link |
They definitely would agree with me that I don't understand it.
link |
What about, is there an understanding of the birth,
link |
the origin, the bang bang? No, no, no, no, there's theories.
link |
There are theories. My favorite is the one called eternal inflation.
link |
The infinity can be on both sides, on one of the sides, and none of the sides.
link |
So what's eternal infinity? Okay.
link |
Infinity on both sides. Oh boy. Yeah, yeah, that's...
link |
Why is that your favorite? Because it's the most just mind blowing?
link |
No, because we want a beginning. No, why do we want a beginning?
link |
I practiced it was a beginning, of course, and practiced it was a
link |
beginning, but could it have been
link |
a random fluctuation in an otherwise infinite time? Maybe.
link |
In any case, the eternal inflation theory,
link |
I think, if correctly understood, would be infinite in both directions.
link |
How do you think about infinity? Oh god.
link |
So, okay, of course you can think about it mathematically.
link |
I just finished this discussion with my friend Sergei Brin.
link |
Yes. How do you think about infinity? I say, well, Sergei Brin is infinitely
link |
rich. How do you test that hypothesis?
link |
Okay, that's such a good line. Right.
link |
Yeah, so there's really no way to visualize some of these things.
link |
Yeah, this is a very good question. Does infinity have any place in
link |
physics? Right. Right, and all I can say is very
link |
So, what do you think of the recent first image of a black hole visualized from
link |
the event horizon telescope? It's an incredible triumph of science.
link |
In itself, the fact that there are black holes which collide is not a surprise,
link |
and they seem to work exactly the way they're supposed to work.
link |
Will we learn a great deal from it? I don't know. I can't.
link |
We might, but the kind of things we'll learn won't really be about black holes.
link |
Why there are black holes in nature of that particular mass scale and why
link |
they're so common may tell us something about the
link |
structure, evolution of structure in the universe.
link |
But I don't think it's going to tell us anything new about black holes.
link |
But it's a triumph in the sense that you go back a hundred years and it was a
link |
continuous development, general relativity, the discovery of
link |
black holes, LIGO, the incredible technology that
link |
It is something that I never would have believed was going to happen
link |
30, 40 years ago. And I think it's a magnificent structure, a
link |
magnificent thing, this evolution of general
link |
relativity, LIGO, high precision, ability to
link |
measure things on a scale of 10 to the minus 21.
link |
So yeah. So you're just astonishing though we
link |
that this path took us to this picture. Is it different?
link |
You've thought a lot about black holes. How did you visualize them
link |
in your mind? And is the picture different than you
link |
utilized it? No, it's simply confirmed.
link |
It's a magnificent triumph to have confirmed a direct observation
link |
that Einstein's theory of gravity at the level of
link |
black hole collisions actually works
link |
is awesome. It is really awesome. You know, I know some of the people who are
link |
involved in that. They're just ordinary people.
link |
And the idea that they could carry this out,
link |
I'm shocked. Yeah, just these little homo sapiens.
link |
Yeah, just these little monkeys got together and took a picture of
link |
slightly advanced limers, I think. What kind of questions
link |
can science not currently answer but you hope might be able to soon?
link |
Well, you've already addressed them. What is consciousness, for example?
link |
You think that's within the reach of science? I think it's
link |
somewhat within the reach of science, but I think now I think it's in the hands
link |
of the computer scientists and the neuroscientists.
link |
Not a physicist. Perhaps at some point, but I think
link |
physicists will try to simplify it down to something that they can use their
link |
methods and maybe they're not appropriate. Maybe we simply need
link |
to do more machine learning on bigger scales,
link |
evolve machines. Machines not only that learn but evolve their own
link |
architecture. As a process of learning, evolve an
link |
architecture not under our control, only partially under our control, but under
link |
the control of machine learning. I'll tell you another thing that I find
link |
awesome. You know this Google Bing that they
link |
taught computers how to play chess? Yeah, yeah. Okay.
link |
They taught computers how to play chess, not by teaching them how to play chess,
link |
but just having them play against each other. Against each other.
link |
Self play. Against each other. This is a form of evolution.
link |
These machines evolved. They evolved in intelligence.
link |
They evolved in intelligence without anybody telling them how to do it.
link |
They were not engineered. They just played against each other and got better
link |
and better and better. That makes me think that
link |
machines can evolve intelligence. What exact kind of intelligence? I
link |
don't know. But in understanding that better and
link |
better, maybe we'll get better clues as to what
link |
goes on in our own intelligence. Well, life in intelligence is
link |
last question. What kind of questions can science
link |
not currently answer and may never be able to answer?
link |
Is there intelligence out there that's underlies the whole thing?
link |
You can call in with a G word if you want. I can say
link |
are we a computer simulation with a purpose?
link |
Is there an agent, an intelligent agent that underlies
link |
or is responsible for the whole thing? Does that intelligent agent
link |
satisfy the laws of physics? Does it satisfy the laws of quantum mechanics?
link |
Is it made of atoms and molecules? Yeah, there's a lot of questions.
link |
And I don't see... It seems to me a real question.
link |
It's an answerable question. Well, I don't know if it's answerable.
link |
The questions have to be answerable to be real.
link |
Some philosophers would say that a question is not a question
link |
unless it's answerable. This question doesn't seem to me
link |
answerable by any known method, but it seems to me real.
link |
There's no better place to end. Leonard, thank you so much for talking today.