[MUSIC PLAYING] SANDERS KLEINFELD: Hi, everyone. My name is Sanders Kleinfeld. And I'm excited to welcome
Carlo Rovelli to Talks at Google today. Carlo is a theoretical physicist
who directs the quantum gravity group at Centre de Physique
Theorique of X Marseilles Universite. His contributions to
the field of physics include more than 200
scientific articles, as well as two monographs
on loop quantum theory. He's also the author of the best
selling popular science books, "Seven Brief Lessons
On Physics," "Reality is Not What it Seems,"
and "The Order of Time," which have been translated
into more than 40 languages. He's here today to discuss his
latest book, "Helgoland," which explores the fundamental
principles, challenges, and mysteries of quantum
theory, tracing its genesis back to the revolutionary
contributions of scientists including Niels Bohr, Wolfgang
Pauli, Max Born, Erwin Schrodinger, and
Werner Heisenberg, whose groundbreaking discoveries
on the island of Helgoland helped kick start the
field of quantum mechanics. Thanks so much for
joining us, Carlo. CARLO ROVELLI: Thank
you very much, Sanders. It's a pleasure to be here. SANDERS KLEINFELD: Great, great. So to kick things off,
I have to confess, I've always found the
field of quantum mechanics to be a little bit intimidating. So as I was reading the
introduction to the book it was really comforting
to see you characterize it as difficult to understand. And I was wondering,
for those of us like me who are fairly
new to quantum mechanics, can you give us a primer of
what the field is all about and why it can be
so counterintuitive? CARLO ROVELLI: Sure. So the key point is that
quantum mechanics is weird. It's incredibly weird. So the sense of unease
that you are referring to is not just because it's
advanced, it's science, it's complicated. It's not just that. It's weird for
everybody, including for those who work into it. So let me say in a
couple words what it is, what's good about it,
and why it's weird. What it is is possibly the
greatest revolution in science ever, certainly one
of the greatest. It started 100 years ago, 1925. So it was a century. So it's not something new. And it has replaced
the basis of mechanics of classical mechanics. So it's really the core of
modern science, modern physics. And it's the basis of plenty
of current technology. And we're using a
computer to communicate. Computers are made by little
electronic circuits, which are computed and designed using
quantum mechanics and work thanks to quantum phenomena,
small quantum things. Lasers are a quantum effect. Of course, nuclear
power and nuclear bomb. Plus, it's used
today for justifying the basis of chemistry,
for understanding how the sun works, why birds,
to the formation of galaxies. So it's really the
core of science. And it works fantastically. It works perfectly. In fact, one can say it's the
best scientific theory we have had ever, because we don't have
any single indication of where it goes wrong. It has always went right. Everything it's
predicted is correct. And maybe it's not
the final story, but we have no sense of
where, when are its limits. So that's a goodie. What's the problem? The problem is that
you can use it. You can be an engineering,
a chemist, or somebody studying galaxies, and use it. And a good student-- I mean, students go to
classes at a university and take a course in
quantum mechanics. You do the exercises. You predict. You go to the lab. You do a little experiment. It works out exactly
as the theory predicts. But if you think about,
if you stop and say, wait a moment, what
am I doing exactly? The way the theory
is formulated is completely weird, because
the theory is formulated not telling you what happens,
but only what you observe, what you see. So its like you
treat any system, whether it's a part of
the sun or a little chip in your computer, you
treat it as a closed book, where you look at
it, and you see something. And then you look at it again,
and you see something else. And in between, it
doesn't make sense to fit. If you try to fit
what is in between, you start saying funny things. So the way it's
taught in books is that it is about observations. But what does nature know
whether we are observing or not? I mean, is that about us? And it's very confusing. And people came out
all sort of ideas to understand what
is really going on. And in a century, we are
really in a conceptual mess. We don't know what
it's telling us, the best theory we
have about the world. SANDERS KLEINFELD: Yeah, yeah. So you touched on
this a minute ago. But you describe
quantum mechanics as an interplay of three key
principles, observations, probability, and granularity. I was wondering if you
could elaborate a bit more on the significance of
each of these concepts in the context of
quantum theory, and how their synthesis
allows us to model the behavior of particles. CARLO ROVELLI: Yeah,
so two are easy. And the third one
is complicated. [LAUGHS] SANDERS KLEINFELD: Cool. CARLO ROVELLI: The first you
mentioned is a complicated one. So I'll maybe start
from the easy ones. The easy one's the granularity. And that's pretty easy. I mean, that's just a
discovery about nature. Things are granular
at small scale. They literally are. For instance, light that comes
from that lamp on to my face, I give a description,
electromagnetic wave that comes from me. I see a sort of continuous thing
making me shining and lighting. But if we make a
precise measurement, we see that it arrives on my
face in little photons, grains, granular. So light will interact
with me dot by dot. And, in fact, if we tune
down the light in a screen, we actually see the
dots one by one. And this granularity,
it's all over. I mean, for instance,
the energy of the atoms, it's quantized so that
it's a quanta of energy, grains of energy, angular
momentum for anything. You just discrete values
instead of continuous values. So there seems to be
a sort of bottom level where you go down
to discreetness instead of continuity. I used to say that continuity
is an approximation to many. Infinity is an
approximation to many. Now there is always a
finite number, so to say. That's the easy
thing, number one. The easiest thing, number two,
is that it came as a surprise, but it came as a
surprise just because we were used to Newtonian science. It was discovered with quantum
mechanics that we cannot predict the future exactly, even
if we have maximal information about something, right? So you are taught
in physics classes before quantum mechanics,
when you go to university, that in principle, if
we knew the position and velocity of every
particle in the universe, we could compute
what happened next. Which is sort of true because
things are complicated. But in principle,
it's true that if you had total decision on a
classic physics system, the old idea was the
world is deterministic. So that determines the
future evolution forever. Well, that's wrong. We have discovered it's a
fact that even if you measure something maximally
that you can measure, you cannot say what's
going to happen next. What you can say,
it's a probability of different happenings. SANDERS KLEINFELD: Right. CARLO ROVELLI: And that,
usually it's good enough. Because, first of all,
when you have many things, probabilities just average up. And it becomes
certainty, averages essentially in a certain way. But as soon as you
go into the small, you have this
intrinsic probability, which is a de facto
impossibility for us to predict the future. All right, so that's
the second discovery. So the world is granular
and is probabilistic, cannot predict the future. So far, so good. And then there's a third
one, which is the hard one, and is the one that has given
all the confusion, well, technically it's
called contextuality. And it's a fact that
the theory is not about what happens, but
is about what you observe. Even the language, when you open
the books of quantum mechanics, it's terms of observables
instead of variables. And the theory doesn't say,
imagine you have a pen moving. The theory is not that in every
moment the pen has a position, has a velocity,
has an orientation. It doesn't talk about that. It talks you, if you
look at it, it's here. Then you close your eyes. You look at it again. It tells you where it is. But if you try to reconstruct
what happened in between, the theory doesn't say anything. It's not only that it
doesn't say anything, but if you think about it,
any possible construction in between seems contradictory. And the prototypical
example, Feynman said, all the mystery of quantum
mechanics is this example. Imagine, as I said, you have
something going from here to here. Imagine there is a wall
in between with two holes. It's called the
double slit experiment in quantum mechanics. It's always there. So there's a hole
here and a hole here. So we see a pen in one point. You see a pen in
the other point. And then you say, well,
it has to go through one hole or the other hole. Now, it turns out
that if you assume that it went through this hole,
you'll get the wrong answer. If you assume it went
through this hole, you get the wrong answer. So you have to
assume, so to say, that it went
through both, namely that when you look
at it, it's a pen. But when you don't look at it,
it's a sort of wave all over. And that's weird. That's the weird part of it. So discreetness,
things are granular. Probability, we cannot
predict the future. And observables. The theory, it's not about what
happens, but what we observe. And the third one is a source
of the mystery, which people are starting in
various directions to try to make sense of it. SANDERS KLEINFELD: Right, right. And that third one I
guess is the one that is most famously presented in
the Schrodinger's cat thought experiment that you talk
about quite a bit in the book, and has also achieved a lot of
prominence in popular culture. It's been in movies
and television shows. And until I really
read your book, I don't think I fully grasped
the implications of it. I was thinking about it as
a very epistemological type thing, where you don't know
whether the cat is asleep or awake inside the box. But it's actually really
a metaphysical statement. And I was wondering
if you could talk through the
experiment a bit more and discuss some of
the significance of it. CARLO ROVELLI:
Yeah, it's exactly like the two holes, right? So if you say, well, I
see the particle here. I see the particle on the
other side of the wall, and there are two holes. So I can make a
computation and say, where I should expect the particle. And I get it right if
I do quantum mechanics. But then I say, well, imagine
that in the middle of the time it went through one hole, OK? And then I get it wrong. And then I say, well, OK, maybe
it went through the other hole, and I also get it wrong. And then I say, well, maybe
it went half probability here, half probability here,
I also get it wrong. So the only possibility is
that when I'm not looking, I have to think that
it is in both, right? And then people say, OK,
fine, I mean, pen [INAUDIBLE] in a wave, and then through
hole, and then converge. That's a solution. But that's not a
solution, because if I look which hole it
goes through, if I look I only see it here, and see
it there, or see it there, OK? And then what happens is
that since I've seen it here, the result later is different. So this is actually done
with a particle with two holes in any laboratory. Many universities do
it for the students. And you actually
see the mystery. But if you think instead of a
particle going through holes, a cat having two possibilities. So the whole Schrodinger version
was the cat was dead or alive, but it's not very nice. So if we think the cat
could be asleep or awake, the theory forces
you to say that it's neither awake nor
asleep until you look. And then we're going
to correct that. Because I think that's a
wrong take on the story. Now, what it means exactly? It means that if you assume that
at some point the cat is awake, then you can predict
something, and it's wrong. If you assume that it is
asleep, you can predict a thing, and it's wrong. You really need two,
both in some sense there to make the right predictions. So the theory seems to say that
when you don't look at things, they are in funny
superpositions, that here and also here, the
cat asleep and also awake. And that's not clear,
because Sanders, if you were in a superposition
how would you feel? Right, I mean,
what does it mean? SANDERS KLEINFELD: Yeah. And I think that brings
us to a key question that you ask in the book
where your observations are material in the science. And you ask, why does
nature care whether we observe something or not? CARLO ROVELLI: Exactly. SANDERS KLEINFELD: And
I was curious if you could elaborate a bit on that. CARLO ROVELLI: Yeah,
and this is somehow-- let me now shift to the general
agreed sort of description of the weirdness of
quantum mechanics, to one particular way of trying
to come out of this puzzle. And all the ways to come out
of the puzzle are strange. So this is the one which
I find less implausible. I mean, the other is worse. And it's a fascinating way. It's called the
relational interpretation of quantum mechanics. I've worked on it. Other people have worked on it. Scientists have worked on it. Philosophers have worked on it. So the idea is the following,
that this has nothing to do with observations. I mean, the observation is
the wrong language here. That we observe, it's
the wrong language here. The point, the mistake,
it's confusing relations with observations,
relationality with subjectivity. And to explain
what I mean, think of another case in which we use
the language of observation, but it's nothing to do with
observation in physics, in fact, when we
talk about velocity. Now, I mean, I
think everybody who has taken elementary
field, a physics class, knows that velocity is
relative to the observer. SANDERS KLEINFELD: Right. CARLO ROVELLI: So if
you are on a train, you say, oh, this
has a velocity 0. It's not moving. Meaning it's not moving
with respect to train. But it's moving with
respect to the Earth. So the velocity of something
with respect to the train is different than the velocity
with respect to the Earth. And is different, the velocity
with respect to the sun. And is different velocity
with respect to the galaxy, and so on and so forth. So it makes no
sense to say, what is the velocity of the moon? Well, it depends. It depends with
respect to what, right? So velocity is a
relational concept. It refers to two
things, not to one. And we sometimes don't
say with respect to what, because we give it for granted. I mean, usually we say
with respect to the Earth. We mean velocity with
respect to the Earth. I mean, if you get
a ticket speeding, it doesn't say it goes
that speed with respect to the Earth. But that's what it means. And if a mother on the
train says to the kid, stop, don't move, she doesn't
mean jump out of the window or don't move with
respect to the Earth. It means, don't move with
respect to the train. SANDERS KLEINFELD: Yeah. CARLO ROVELLI: All right, but
obviously the relational aspect of velocity has nothing
to do with subjectivity. There is not an observer here. The moon has a velocity
with respect to the Earth, and velocity with
respect to the sun, but not because the sun
observes the moon, has a brain, has a consciousness. It's nothing to do
with observation. It's just relations. So I and others think
that the correct way of thinking about
quantum mechanics is that it tells us that all
quantities are like velocity. So they are really relations. All physical variables are not
characteristic of the object itself. A characteristic
describes how their object interacts with something else. So things have properties
relative to something else with which it is interacting. And the property describes
the interaction, not what happens to the object itself. So objects by themselves
have no properties. They have properties
relative to other things, and these properties
mean the interaction. Now, if you take
that perspective, you throw away the idea of
observers, subjectivity, we-- nature doesn't care if
we look or not at things. But nature does care if
two things interact or not. That's a fact of nature. SANDERS KLEINFELD: Right. CARLO ROVELLI: So when the
electron hits the screen, that's a fact. And the electron has a position
with respect to the screen. OK? If the electron
bumps against a wall, it has a position with
respect to whatever it has bounced around. But the subtlety-- and that's
the way out of the tunnel-- is that if something
has a value-- the variable something
has a value with respect to something. This has no influence of
the value of that variable with respect to something else. So the cat. In the little Schrodinger
story of the cat, there is a little
poison or something that makes the cat sleep
that might or might not open inside the box. So does it open
or does not open? Well, with respect to the cat,
obviously it's one of the two, not both. Because it's an actual thing
that happened with respect to the cat. But what happened with
respect to the cat doesn't imply anything
definitive with respect to me, until I interact with the cat. So the fact that the poison
has got to the cat or not, it's a fact with
respect to the cat. But this doesn't forbid
the fact that if I want to compute something
with respect to me, I have to consider both
possibilities until I do the calculation. And that's the way out. So the way out is relationality. That's why it's called
relational interpretation. Think about reality
not as a collection of things that have properties,
but a collection of objects that interact with one another. And the properties come
out in the interactions. So think relations rather
than individual objects or properties. It's a relational thinking. Which is not new, in a sense. Because we think relationally
plenty of things in our life. A lot of things are relations
in our life, are not entities. It's surprising that
somehow all the way down to elementary physics, it's
relations all the way down. SANDERS KLEINFELD: Yeah, yeah. It's a real key paradigm
shift of quantum mechanics, a shift away from matter
as the essence of reality. And in the book I think you
expressed it really eloquently. You said, there are
no elementary entities that we can describe
except in the context of their interaction
with something else. This leaves us without a
foothold, no place to stand. If matter was definite
and univocal properties does not constitute the
elementary substance of the world. And if the subject
of our knowledge is a part of nature, what is the
world's elementary substance? From where can we begin? What is fundamental? And I guess that's
the question that I'd like to throw back at you. If we're only in a
world of relations and relational phenomena,
what is fundamental? I think as humans, we are
looking for something really substantial to plant a flag on. CARLO ROVELLI: You
want a sincere answer? SANDERS KLEINFELD: Yeah. CARLO ROVELLI: I think we make
a mistake by aiming there. You said we as humans aim
at some basis, something fundamental from which we
derive everything else. If there was one, we would
have found it already. [CHUCKLES] You know, humankind-- let
me just open up the topic a little bit. Humankind has been looking for
the fundamental [INAUDIBLE] from which everything
else derives since ever, with a long
list of attempts, right? From God, to matter,
to phenomenology, observations, [INAUDIBLE]
circles, whatever information, energy. At some point people
say, oh, it's all energy. It's not matter. Philosophy, it's
a constant attempt to find the principle from
which everything else derives. And nobody has been able
to convince anybody else. But [INAUDIBLE]. Why? Because first of all, you can
start from different points and reconstruct the
world in different ways. The world admits
to be reconstructed in different ways. But more importantly, the
world is a complicated set of things which we can
understand a lot about. We do, because
science is incredibly powerful in connecting
things, right? But physics is very
good, because it connects to everything. There's a sense that we
expect that everything satisfies physical laws, right? So the universality of
physics remains very strong. But physics has tried repeatedly
to find the basic ingredient. And what is the basic
ingredient in modern physics? In quantum field theory and
general relativity, and quantum gravity, it's very slippery. SANDERS KLEINFELD: Right. CARLO ROVELLI:
Because the particles, the quarks, the
electrons, they're not the basic ingredients. They are created, destroyed. They are quantum
excitations of a field. They are all these
granular things that depend on the
way you look at them. In quantum mechanics
come like a violent bomb there and shows that
it doesn't work. We don't have a clear ground. But that does not mean that
we don't have a good way of thinking about the world. We do have plenty of good ways
to think about the world, which are coherent with one another. So let me say this. This is a pen, right? And there seems to be
just plenty of properties of the pen itself. It's a pen. It's black. It's oriented in this way, in
this position, this velocity. It's made by atoms,
whatever you want. It's this color. Now, if we look at all
these properties one by one, we start seeing that
they're not in the pen. For instance, the
fact of being a pen obviously is because
we write with it. Somebody came from Andromeda
and looked at this thing says, well, it's a thing,
but it's not a pen. Because a pen is
something you write on. So the fact of being a pen
depends on our use of it. It's black. But that's a color. The color depends on our vision. Animals see things
with different colors. So the color is
really something that pertains to the way my
eyes work, the light works, bounces on the thing. All right, so it's
a single piece. Well, it's not true. I can take it apart. It's made by atoms. Each one has its own
position and velocity. Well, it's not true, because
quantum mechanics tell us that the atoms, the
position and velocity depend on the interaction with me. Does this take away
the reality of the pen? No, the pen is very real. But it's a combination
of all these relations it has with everything else. And if you think for
a moment, suppose it didn't have any relation
with anything else, I couldn't see it. I couldn't touch it. I couldn't do anything with it. Or nobody else. So what would it
mean to be real? Reality is a strange notion. We can play around. So I think that the
shift, you said, is a major shift,
because we tend to say, OK, whatever reality
is, it's a set of things with properties. The shift from that to reality
is a set of interactions, and the objects are the
nodes of this interaction, a pen is a node of all
these possible interactions, it's not dramatic in
our everyday thinking, because we think in
terms of interactions. But it's dramatic in our
foundation of thinking. And I think it's a step ahead. It's not a step back. SANDERS KLEINFELD: Right, right. Fundamentally,
everything is relational. And as you were also
alluding to earlier, there's a probabilistic
element to it as well. And I was interested if you
could elaborate a little bit more on that as well. It's sort of the quintessential
question that Einstein posed. Does God play dice? Or as you rephrased it in your
book, are the laws of nature really not deterministic? And in the book you say,
100 years after Heisenberg's and Schrodinger's
bickering on this, the question is still open. So I'm wondering what the
implications are, in your mind, in terms of the deterministic
nature of everything that we experience in the world. CARLO ROVELLI: Yep, absolutely. It's true that we're
still bickering about it. I have colleagues,
in fact, that do not like this relational
reading of quantum theory. And one of the reasons for
which they don't like it is because they're very
attached of the idea that the world is deterministic. So whatever happened
today determined what happens tomorrow. And, in fact, there
are other attempts to make sense of quantum theory. But careful. Let me be precise here. I keep thinking
about these things. Because in these other ways you
assume that behind the scenes, so to say, there
is something else which is going on which
is really deterministic. But for one way or the other,
we have no access to this, or we cannot predict it. SANDERS KLEINFELD: Right. CARLO ROVELLI: Let me
just make it concrete. Because there are two
alternative interpretations of quantum mechanics which
are the most popular ones besides relational one. One is the hidden variables. And as the name of the
interpretation implies, there are some variables
which are hidden, OK? So something happens
that is, in principle, impossible to access to,
which determines what's going to happen tomorrow. But it doesn't change the
fact that we cannot predict the future, right? Precisely because these
variables are hidden, it remains the fact
that we cannot. So if you have something
you don't know tomorrow, you can always imagine there
is something today that determines it, and you
don't have access to it. You can add to
your set of beliefs of what exists that the
information of tomorrow is already there today. But what do we get? And another popular
interpretation of quantum mechanics, which has also got
out in the popular press-- people have written
books about it-- it's the so-called many
worlds interpretation, in which the cat is really
awake and also asleep. Well, when you
look at it you see only one, because you
yourself split in two. And so there is one version of
you that sees the cat awake. And one version
sees the cat asleep. So nothing is
probabilistic here, because there are two of you. But it is
probabilistic, the fact that you don't know
which one you are going to be tomorrow, right? Because I look at the cat. I see it awake. And I say, OK, of
the two, there's my brother who has
seen-- but I didn't know before which one I've got. So I cannot make a prediction,
because I don't know in which sense I can make it. So again, I have
to be [INAUDIBLE].. So it doesn't really matter. That's where I'm going. Whether nature's probabilistic
or not with respect to God who sees everything,
assuming that she exists, she, God, exists. In a sense, even if the world
was completely probabilistic, you would say, OK, well,
God sees the future. So it's already determined. So the question is
not so deep after all. I mean, just things happen. And it's a fact that it's
impossible to predict what's going to happen tomorrow, given
maximal available information today. So I think that at the end of
the day, the unpredictability which has been discovered
in fundamental physics, it's just a fact we
have to confront. And there's nothing
wrong with it, right? I mean, human kind has
lived for millennia thinking that the
future is unpredictable. I don't know why we should
be attached to the idea that there is absolute
determinism or predictivity. SANDERS KLEINFELD:
It's true, true. I think, perhaps,
maybe we just want to be able to
predict the future, and that's all there is there. CARLO ROVELLI: Oh, yeah. We would like so much to
know what happens tomorrow. SANDERS KLEINFELD:
Would be nice. [LAUGHS] CARLO ROVELLI: Of course. Maybe that's all there is. Yeah, right. SANDERS KLEINFELD: Right, right. So I think any discussion
of quantum mechanics would be incomplete if
we didn't at least touch on the concept of entanglement,
which you describe as a phenomenon by which
two distant objects maintain a kind of weird connection,
as if they continue to speak to each other from afar. So you describe this
scenario in which there are two entangled photons,
one that's sent to Vienna and one that's sent to
Beijing, each of which could be like the color
red or the color blue. But when observed, they always
turn out to be the same color. So this is very, very
counterintuitive, certainly for me, and I think for
a lot of other people. How does this work? These connected
particles that could be separated by vast distances? CARLO ROVELLI: It
is counterintuitive. And it's very subtle
when you go into details. And, in fact, it's
a subtle phenomenon. I even hesitated
putting a chapter in the book about that,
because to get it right exactly what happens, it's subtle. But then, as you
see, it's really the sort of quintessential
quantum phenomenon, this strange connectivity. It's no sense in
which you get a sort of sudden transfer of
information from one to the other. So it's not that one of the two
is blue, decides to be blue, and tell the other,
hey, hey, I'm blue. You too. You have to be too. That's not what happens. We are pretty sure about that. Neither what happens is that
there was a previous agreement. So they knew to be
both blue before. And we have good reasons to
exclude that one as well. So it's something
far more subtle. And what is far more
subtle is that-- let me put it in this way. To check that they're both blue
you should be in both places instantaneously. And nobody is in both
places instantaneously. So what do we mean
when we say that there is an instantaneous
transfer of information? Nobody can check that. So what is the sense of that? What we really mean is that
whatever happened there, if we bring the information
up, and then later in time you compare them, at the
moment in which anybody can make this
comparison, it sees that they are both the same. SANDERS KLEINFELD: Right. CARLO ROVELLI: But
since phenomena are relative to
an observer, it's only in the moment in
which the comparison is made that this has happened. Before it's like the cat,
which is both alive-- asleep and awake. So with respect to Beijing,
something happened. With respect to Vienna,
something happened. With respect to whoever gets all
the information and compares, something happened. But with respect to each
there is a coherent story where information never jumps. It's only when you bring
pieces together and try to juxtapose it that you get
this apparent instantaneous transfer of information. It's not real,
because you cannot. You're not allowed to
bring these pieces. Reality is really
relational in this sense. Then this relationality, you
don't see it microscopically, because it's a minimal thing. You have to do super subtle
experiment to bring it up. But the world with respect to
me and the world with respect to you cannot be made
exactly coincides. There are little discrepancies. The perspectival aspect that
forbids a coherence down to the Planck constant scale. SANDERS KLEINFELD: Mhm. Yeah, so you refer to it as
basically a dance for three, I believe, where you have one
particle, the other particle, and then the observer. And we're all participating
in this together. CARLO ROVELLI: Exactly. And that's a way of putting it. Entanglement is a dance
for three, not for two. Because whoever compares
is the third perspective. And the third
perspective is the one that compares, and
is the one that doesn't see the long
distance information jumping. SANDERS KLEINFELD: Right, right. So earlier in the talk
and also in your book, you touched really
briefly on some of the applications
of quantum mechanics that have already
begun to transform technology and society. And you mentioned lasers,
semiconductors, and the physics of formation of galaxies. I was wondering, if you look
10 to 20 years down the road, are there any other
groundbreaking advances you can potentially
foresee scientists making as a result of all the
research into quantum theory that you're doing? CARLO ROVELLI: People are
working on quantum computers today. There's a huge investment
in quantum computers. You can exploit this
quantum weirdness to make computers
that are immensely faster than the
current computers, with obvious applications. The theory of quantum
computers is absolutely solid. And so it's perfectly
possible to do them. The technology, it's hard. Because an object [INAUDIBLE]
quantum is an isolated object. As soon as it starts to
interact with too many things, its quantum weirdness
disappears, OK? Sort of it's diluted
out in the interaction, and you don't see it. And that's why the actual
cat-- the cat is a big thing. The quantum interference
between the two cats, it's a very delicate phenomenon. You don't see it because
the cat is too big. So you can always assume,
oh, the cat was asleep. Oh, the cat was awake. So big things tend to
hide quantum behavior. And that's why making
a quantum computer is hard, because you
have to isolate it. So I don't know this technology. I'm a theoretician. You have to ask the
people doing technology, if it's actually
doable as people hope, or it is more hard
than one would expect. It certainly would change a lot. I'm sure there are other
applications of quantum mechanics we haven't seen yet. They are going to come up. It takes long to-- general relativity, it's
older than quantum mechanics. It's 100 years old. We have the GPS that works
thanks to general relativity. Nobody would have ever
imagined in Einstein's time that you're going to use
general relativity to find your way to downtown
Manhattan, to find where is the 27th Street. SANDERS KLEINFELD:
Yeah, so we'll just have to stay
tuned I guess, and see what can be developed on
top of quantum mechanics. CARLO ROVELLI: That's right. SANDERS KLEINFELD: Great, great. So we have a bunch of great
questions from the audience. So we're going to
pull them up now. And we have a question
from Timur here. "Are the results of the double
split experiment the same when the observer
is not a human? For example, a trained animal,
a robot, camera, et cetera?" CARLO ROVELLI: Yeah, I'm
very happy of this question, because in a sense,
this is a key question. And the answer is
a resounding yes. A strong yes. I mean, there is
absolutely no evidence that there is anything special
with humans when things happen. You can try do any quantum
experience, interference, or entanglement,
or anything else, and replace the looking
human with a robot. And the result is
absolutely the same. And that's what strongly
indicates that-- There are some people
who say, oh, the quantum mechanics really happen when
consciousness sees something. But consciousness has
nothing to do with that. SANDERS KLEINFELD: Great. JZ is wondering, to build
on Timur's question, "Do the results depend
on the expectation of what the results might be? For example, if expecting
the results to be A or B, do you get a different result
versus if expecting A, B, or C?" Do our expectations matter? CARLO ROVELLI: No, they don't. Certainly they don't. What matters--
don't get it wrong. So listen the full thing. What matters is
what we look at, OK? So if we look at
one thing and if we look at another
thing, the result, it's contradictory
sort of, in one sense. But not because we look at
one thing or the other things. Because the kind of interaction
that we have in one case, and the kind of interaction
that you have in the other case probes the system differently. And the system behaves,
manifests itself in interactions. So where there is an
interaction of one kind, or interaction of
the other kind, that brings up different
properties of the system. And this might be
contradictory to one another. Namely, sort of one
interaction sort of cancelled the information about
the other interaction. And this was completely
clear in the early days of quantum mechanics, when
Bohr, Heisenberg, and company were discussing on that. But they were thinking
in terms, I observe this, and I observe that, right? I use this machine to interact
with the quantum system, and this machine to interact
with the quantum machine. Had nothing to do with the
machine, with the apparatus. It's just the kind
of interaction, the kind of Newtonian
interaction term that is in play that
determines which quantities are realized in an interaction. So nothing human, nothing
depending on what we expect, or anything which is mental. Everything is down
there physical, who is interacting with whom. And the properties are in the
interaction, not in the things. SANDERS KLEINFELD:
Another question. David is wondering,
"This relationality seems similar to the fact that
we can't interact with anything outside of our light
cone in relativity or the inside of a
black hole in general. Do you view this the same way? CARLO ROVELLI: Yeah,
there is a quite analogy, even if there are differences. Because the theories
are different. And the details are different. But there's a strong analogy. In fact, even more, because
Let me put it this way. Einstein's relativity,
it's a discovery that we will make a mistake when
we assumed that simultaneity is absolute, OK? There are two things
happening in distant places. We can say which
one happened first, or if they happened
at the same time. So there was a wrong assumption. And, in fact, the
remarkable thing is that Einstein
understood that. That's what Einstein understood. In a formula this
was already there, because it had been developed
by Lorentz, by Poincaré, by mathematical
physicists before him. So he understood what it meant. And with quantum
mechanics it's similar. Because we have this formula
that works to describe nature. But we didn't understand
what it means. And relational quantum
mechanics, the idea, yeah, we understand
what it meant. It meant that we're
making a mistake, like relativity we make
a mistake of thinking there is absolute simultaneity. Here we're making a
mistake in thinking that there is an absolute
reality of the properties independent of the interactions. So simultaneity is relative
to a state of motion with respect to
which you define it. The values in
quantum mechanics are relative to the kind of
interactions which are there. And in both cases-- in fact, the language is
strong, because in both cases we usually are-- we use the language
of observers. Relativity is relative
to the observer. And quantum interactions are
relative to the observer. Quantum measurements are
relative to the observer, as if it was
something subjective. But it has nothing to
do with subjectivity. It's relativity in both cases. SANDERS KLEINFELD: A
question from Krish. "If all particles are the result
of fundamental interactions, does it also apply to
the fundamental point particles described by
the standard model too?" CARLO ROVELLI:
Yes, very much so. In fact, I don't know if
Krish is a physicist or not, but the popularization
version of the standard model is that, yeah, there are
certain number of particles, and they bounce
against one another. And that's the world. But if you take a good class
in quantum [INAUDIBLE] theory and try to understand the
quantum model for real, that's not so
simple, not at all. Because these
particles are actually quantum excitations of a field. So the particles
are not down there. The particles are
what happens when the field interacts with you. So much so that
when we compute-- in particle theory, you
throw two particles in, and then you see what comes out. And you compute it before. In the computation,
you're actually summing the different
Feynman diagrams, which are different stories
of what happened in between. So to compute the probability
of some outcome you consider all possible outcomes. And you sum all of them. So that's exactly
Schrodinger's cat. You're saying the two particles
come in, and this happened. But also this happened. But also this happened. So it's like the Schrodinger cat
is alive, but he's also asleep, and he's also this. And he's also this. And he's also this. Or the particle
through the two holes. It passes by here, but
also here, also here. So explicitly, in quantum field
theory, if you look at it, the way we do
calculations is not to add on that the world is
made by particles that bounce. The world is made by particles
that are particle only when they interact with
me, or with an apparatus, or with something else. In between, they open up
in a cloud of possibility exactly as I was saying before. So yes, the particles
of the standard model are not entities. Let me be precise. I mean, Google is
full of smart people. So let me be precise. [LAUGHS] In fact, I'm giving
this answer to you trying in fact to be much more
sharp and precise than usual. When I talk to the
general public, I'm assuming that you don't
know quantum mechanics. But still, particles
are not entities in current standard model,
quantum field theory. Particles are modes of
interactions with the field. It's how the field
interacts with something else, my machine, my detector
at CERN that measures something. SANDERS KLEINFELD:
Another question. Sanghamitra is wondering,
"Doesn't interaction change the state of what's
being interacted with? So is there any way to actually
know the true nature, state, attributes of anything
in the quantum world?" CARLO ROVELLI: The answer to
the first question is yes. And, in fact, it was
Heisenberg himself, which is the true
person who opened the quantum mechanic door
first who a few years later noticed it. So in quantum mechanics, when
you interact with something, you always affect it somewhere. And for a while it was hoped-- in fact, he hoped
that was it, that was the only thing
we have discovered. The only thing we
have discovered is that there is no way
to affect something, to measure something without
changing it a little bit. But that's not sufficient
for understanding quantum mechanics. [INAUDIBLE] with this
discussion clarify it. This is true. You cannot delicately look
and not affect the system, except in some particular
situations which you can. But that's not sufficient
to explain the mysteries of quantum mechanics. Now, for the second
part of the question, again, I think that the
right direction to think here is not, all right, so
we affect what is there so we don't know what is there. I mean, forget
what we don't know. We are not describing
an unknowable world. If we try to fill up, we
get strange nonsense stories which are useless, in my
opinion, and misleading. We should describe
the phenomena. And the phenomena is
what happens interaction, not what happened
before or after. And that's the key message. Talk in terms of
relations, not what is-- And if I may, just a
small parenthesis there, to enlarge a little
bit the discussion, we think in terms of
relations all the time, right? People who do software, they're
not dealing with entities. They're dealing with relations. People who do
psychology, they're not doing with entities. They're doing with relation. People who do economy, they're
not really doing with entities. Entities are so far away. They do it with relations
between economical actors. If you think how your family is
structured, that's relational. Now, it's not entities. We use relational
notions all over. And when I talked
about the pen, I clarified that so many things
we consider in the things are really relational. [INAUDIBLE] message is that
if quantum mechanics can be thought of
effectively in this way, we can look at our general
way of looking at the world and ask, let's get away
with this obstination of we need to find the entities
behind any phenomenon. SANDERS KLEINFELD: Question
from Beau on the many worlds hypothesis. If we observe the cat asleep,
wave function has collapsed. Is this because we'll exist in
the reality with this outcome versus a separate parallel
one where the cat is observed awake? CARLO ROVELLI: So
let me distinguish. In the many world there is no
collapse of the wave function. So the wave function
never collapses. The wave function has two
components, one with the cat asleep, one with
the cat not asleep. I look at the cat. And my wave function
splits in two. So now there are two Carlos,
one that sees the cat asleep. One sees the cat sleeping. And it's all. So this means that there
are not two copies of me. There are plenty of
copies of me that see the world in all possible ways. This is coherent. It's not incoherent. I have intelligent friends
that think this is reasonable. But I don't think
this is useful. It's multiplying copies
of all of us in a way that I feel doesn't help. From the relational perspective,
there is no wave function. There is nothing to
collapse because there is no wave function. Wave function is just
our calculations, like the prediction
of the weather. You look at the thermometer,
and you change your prediction about the weather. But it's not because
the weather changes. You get some more information,
and your prediction jumps. What is real, or
what we want to call real in the relational
interpretation is just the actual seeing the
cat alive or asleep, or not, or some interference
fact of the cat. So again, there is no
collapse of the wave function in the world. There is a collapse
in your calculations just before you churn
your data, because now you want to compute something else
whose probability's changed because something else happened. SANDERS KLEINFELD: Next
question from Joao. "Thanks for the talk. How do you suggest
combating the idea that quantum mechanics needs
no interpretation and those who are against the study of
the foundations of physics?" CARLO ROVELLI:
Good, so there are two arguments for
quantum mechanics needs no interpretation. One is good and one is bad. So one is good is that,
well, I don't care. I use quantum
mechanics, and it works. And that's a very good argument,
because if an engineering is studying quantum
computers, say, I don't care about interpretation. I know exactly
what my computer is going to spit up, spit out,
the probability distribution of the right outcome. And I don't want to know
what happened inside. He's right. He doesn't need to think
too much about that. But science is not only
about applications. Science is about understanding
better, and going ahead. I do quantum gravity. I want to apply
quantum mechanics to general relativity. And I need to understand
quantum mechanics better to understand how to apply
to general relativity. The objection-- the
second argument is, well, the different interpretations
cannot be empirically distinguished. At present, relational quantum
mechanics, many worlds, hidden variables, you
cannot make an experiment to distinguish it. So what are we talking about? And here is my answer to that. Is the Earth the center
of the universe or not? That was a huge debate
in the Renaissance, all the way from Copernicus
to Galileo, Newton. You might say, of course,
it's a scientific debate. Not only is it a
scientific debate, it was an extraordinary
fruitful scientific debate. Because changing our
views of whether or not the Earth is the center of the
universe completely changed our way of thinking
about motion, about velocity, about
the solar system. We rearranged things. Instead of Earth, planets, we
had sun, planets, satellites. So it's completely rearranging. This rearranging was needed
to go to do Newtonian physics, to do physics. But if you think for
a moment, the Earth being the center of the universe
is a testable hypothesis? Can we measure it, whether
the Earth is the center of the universe or not? No, of course, we cannot, right? There's no measurement that
tell us that it's the center-- In fact, many people at the
time, including the people who condemned Galileo, by
the way, were saying, what are we talking about? The distinction between
the Copernican system, when the Earth spins,
and goes around the sun, and the old Ptolemaic system
with Earth in the middle cannot be solved empirically. And they were right. It cannot be solved empirically. Because you can
always rearrange this from the point of
view of the Earth. It just everything
becomes more complicated if you don't go on any science. So science is not
just about things which are distinguishable
empirically. Of course, science needs to
check theories and distinguish them empirically. But the actual discussion is
how about to think the world? What is the best perspective
to think the world? And the Copernican revolution,
the Copernical perspective obviously was better. History clarified
that completely. We would not have
had the Galileo high against Kepler and
Newton in modern science without the
Copernican revolution. So it's a similar
discussion, I believe. It is a philosophical
discussion. It's a conceptual
discussion, but it's a scientific discussion. There is one perspective that
I think will take us ahead in the future. And I have my own views on that. Others have different views. But the actual discussion
is very productive. And it's core of the
scientific debate. So if a scientist wants
to ignore it, fine. Do something. We don't have all to
do the same thing. But if the argument is this is
not a scientific discussion, that's definitely a wrong
argument, in my opinion. SANDERS KLEINFELD: Next
question from Doug. "Were you surprised
to find echoes of these ideas in Buddhist
philosopher Nagarjuna's writings? Have his ideas changed or
influenced your thinking about quantum mechanics?" I guess someone's read the book. CARLO ROVELLI:
Somebody read the book. Yeah, the book-- in fact, we've
been talking about my book. The book is a lot about
quantum mechanics. It's not just this
part that we discussed, sort of the key ideas. It's also about a description
of the historical origin. So Heisenberg discovered
the theory in the island. And Schrodinger
discovered the theory while making love with
his secret girlfriend, and all that. But then there's a long part
of the book in which I discuss a sort of more general
philosophical implication, implication of this
relational perspective. And one chapter I have,
Lenin, Lenin philosopher, the head of the Soviet Union
in his philosophical texts discussing with a great
Russian intellectual, which is Alexander Bogdanov,
of issues which are exactly parallel to the discussion
about quantum mechanics. And then I have a chapter
of the book about Nagarjuna, which is a Buddhist
monk many centuries ago, second century in India,
which is a super classic of Indian philosophy. It's like Aristotle or Plato
for Western philosophy. It did not influence me
in thinking about quantum mechanics originally. I discovered it very late. And I discovered it because
people kept telling me, this relational perspective
that you are defending, it resonates a lot with the
perspective of Nagarjuna. So I read it, and I was
extraordinarily taken by this. Not the mystical aspect of this. This is not a mystical reading
of the Book of Nagarjuna. As a logician, Nagarjuna,
and as a metaphysician, it has a way of describing
reality in terms of relations, exactly. So in a sense, it
offers a rather solid philosophical
perspective in terms of which you can think of the
relational interpretation. You don't have to think of
the relation interpretation in terms of Nagarjuna thinking. But it has strong
arguments, for instance, counter objections like, oh, you
cannot have relational without having entity with
properties to start with. And Nagarjuna is very clear. He says, that's wrong. In fact, it's the opposite. If you have a relational,
it's the same really. The idea you cannot build
relations out of entities with properties. There is something
primary in relations. So it did help me
clarifying things, and discard some wrong idea. I took something out
of it, definitely. But it was not influential
in actually coming out in me and my colleagues ideas
of relational interpretation of quantum mechanics. And obviously,
Nagarjuna knew nothing about quantum
mechanics, obviously. [LAUGHS] SANDERS KLEINFELD:
Great, so we have time for I think one more question. CARLO ROVELLI: Oh, time flies. SANDERS KLEINFELD: Yeah. A question from Brian. "How do you view
time and a series of instantaneous observations
within the framework of relationships?" CARLO ROVELLI: That's
the last question? It would require
a long discussion. I could give you my perspective,
but it would take time. So let me say something else. Quantum mechanics was
born in 1925, not in 1926 when Schrodinger wrote
the Schrodinger equation, the year before. And in fact, Heisenberg
got the Nobel Prize for the invention
of quantum mechanics with a spectacular
paper, which he just got the right idea of
observables and no [INAUDIBLE].. But then that was very,
very, very confused. I was Max Bohr with
his collaborators who actually put up the theory. So the full quantum mechanics is
in two or three papers written by Max Born and
his collaborator, and Heisenberg,
Paul Dirac in 1925. And what Max Bohr had in mind-- and I think he was on the right
track about space and time-- is that when we think
about quantum phenomenon it's wrong to sort of presuppose
a continuous space, positions, and time, and put into
this space the particles and try to localize everything. It's the other way around. We should think of
quantum mechanics as a set of
interactions, I would say today, observations, a set
of facts which are discrete. And the temporal,
spatial reality comes up from the large number
of little facts happening. So there's a discreetness
that builds up the continuity of space and
time because we have large scale with respect to the
Planck constant. That was Max Bohr's perspective. And I think that's the
right way of viewing that. So we should start from the
realization of variables and interactions in standard
quantum mechanics, the outcome from the measurements in
the language of the books, and not put space
and time before that. We should put space
and time after that. And that becomes a bit
confused in saying like that, but it becomes very, very
clear in quantum gravity. Because in quantum gravity,
that's exactly the case. There's just no
space time continuum. Space time continuum is built
by the quantum phenomenon. SANDERS KLEINFELD: Well, on
that note, we are out of time. So I want to thank you
again, Carlo, for joining us at Talks at Google. And everyone, please check
out Carlo's new book, "Helgoland," now available. So thank you again
very much, Carlo. CARLO ROVELLI:
Thank you very much. This was great. And thank you for the
questions, and also for the great questions
of the public. Very, very sharp and good. I enjoyed it a lot. SANDERS KLEINFELD:
Great, thank you. [MUSIC PLAYING]
