Could there be more than one theory of everything?

Daniel talks to philosopher Thomas Barrett about whether there could be multiple effective descriptions of our Universe 

See omnystudio.com/listener for privacy information.

Episode Transcript

Speaker 1 (00:08):
As I sit down to record today's episode, the beams
of the Large Hadron Collider are powering up. Today will
be the first day of collisions in twenty twenty three,
after a long shutdown to rebuild and upgrade our equipment.
We hope to see some new particles, to discover some
new forces, to reveal new truths about the nature of

(00:29):
space and time and matter and energy. But what if
everything we're learning is not really the truth? What if
it's just a story we tell ourselves. What if there
are other possible stories that could also explain what we
see in our experiments? How would we ever even know?

Speaker 2 (01:04):
Hi?

Speaker 1 (01:04):
I'm Daniel. I'm a particle physicist and a professor at
UC Irvine and something of an amateur philosopher of science.
Welcome to the podcast Daniel and Jorge Explain the Universe,
a production of iHeartRadio in which normally we dig into
questions about how the universe works, what's going on the
tiny quantum level, what's at the heart of black holes?

(01:26):
How does everything work, and how does it weave itself
together to make the incredible, beautiful, mysterious reality that we experience.
My normal co host and friend Jorge can't be with
us today. So I'm going to take the opportunity to
do something a tiny bit different and do a bit
of a dive into some questions in the philosophy of science,
because I really want to understand not just what's happening

(01:48):
in the universe, but why it's happening. What are we
really learning about the universe when we do particle physics,
or when we do any kind of science. Let's make
it concrete. Example, when we build the Large Hadron Collider,
we did it because we wanted to uncover how the
universe works. We want to know answers to questions like

(02:09):
is space filled with quantum fields that buzz and wiggle
to make particles? How many of those fields are there?
And what are the rules of their interactions? Is there
a field for a dark matter? Are there other fields
that we never even imagined? Are all of the fields
really part of one superfield in a way we haven't
ever thought of. These are the kind of questions we

(02:30):
think about, and the language of physics these days is
always the language of these fields. We imagine that space
is filled with these quantum fields, and we've been very
successful at using these quantum fields to describe everything we've
seen so far in particle physics. But there's always a butt.
But nobody has ever actually seen a field. What we

(02:51):
see in the collider are sprays of particles that are
the debris of collisions. The collision itself happens too quickly
and is too small to actually look at, and even
the particles that we do see don't prove that fields
themselves actually exists. When a particle bends in a magnetic field,
what we see is the particle bending, not the field itself.

(03:15):
There's always this part of the story where the field
hides behind a curtain. It's an unobservable part of our
science that helps explain what we do see, but is
never itself directly observable, And that unobservable part, in my opinion,
is really important. It's the juicy bit. It's the real
explanation for what's happening out there in the universe. It's

(03:37):
how we think about physics. We hope, or we wonder,
at least, whether the universe is also doing its calculations
to decide whether a particle goes this way or that
way using fields. I mean, are those fields really out
there pushing and pulling on stuff when we're not looking.
Or are those fields just our description? Are they part

(03:57):
of the story we are telling ourselves about the particles?
Are those fields just part of our minds? Are they
really deep elements of the universe itself? One really fun
way to probe this famous question is to wonder, could
there be another explanation that's just as good? Is there
another way to describe the universe that doesn't need fields,

(04:19):
that tells a different story about what's happening out there
in the universe. So today on the podcast, we'll be
asking the question could there be more than one theory
of everything? To help me dive into these thorny questions
of philosophy, I've invited a guest expert, a philosopher of

(04:41):
science who specializes in this very topic. Okay, so then
it's my great pleasure to welcome to the podcast Professor
Thomas William Barrett. He's professor at the University of California
at Santa Barbara in the Department of f Philosophy, and
he works mostly on philosophy of science and logic, and

(05:04):
on his homepage he has a picture of himself as
an infant in front of a Philosophy Building. Welcome to
the podcast.

Speaker 2 (05:10):
Thanks very much, thanks for having me. It's a pleasure
to be here.

Speaker 1 (05:13):
So does that picture mean that you've been doing philosophy
since you were a little baby?

Speaker 2 (05:17):
In a sense, I guess. I mean everyone tries not to,
but it's, you know, unavoidable. We've all been doing philosophy
since we've been little babies. Well.

Speaker 1 (05:27):
I know people like to say that babies are scientists
because they're all exploring the world and trying to figure
out how things work. But I'd be fascinating if babies
were also philosophers of science at the same time.

Speaker 2 (05:37):
Yeah, I mean, honestly, in that picture, if you look closely,
I have my little truck in my hand. At the time,
I was probably more interested in the truck than in philosophy.
Some days now feel the same.

Speaker 1 (05:49):
All right, Well, before we dig into the topic of
the day, we'd like to get to know you a
little bit about where you stand in the sort of
big questions of philosophy. So let me ask you what
I think is maybe the most important question in philosophy
of science, which is, in your opinion, does a star
Trek transporter actually move your molecules from one place to
another or does it kill you and reassemble you somewhere else?

(06:12):
Is it a murder machine or a teleportation device?

Speaker 2 (06:14):
I don't know, man, your guess is as good as mine.
I'm more of a Star Wars guy than a Star
Trek guy. I honestly, I think I've only seen like
maybe one or two Star Trek episodes. I saw one
of the movies when I was a kid, But I'm
not a Star Trek guy. It's been Star Wars all
the day.

Speaker 1 (06:34):
All right, Well, then, what's the deepest question in philosophy
raised by Star Wars? Is it whether the force actually
has a scientific explanation or if it's just magic?

Speaker 2 (06:45):
Yeah? Science versus religion questions? Yeah, that's good stuff, Way
above my pay grade though.

Speaker 1 (06:54):
All right, Well, we'll try to talk about something a
little bit easier here on the podcast today. What I
want to talk about drills into the question of, like
what are we learning when we are building our theories
of the universe. You know, I'm a particle physicist, and
here in our department we try to or we imagine
that we're trying to build a description of what's happening
in the universe that when we draw a little finement diagrams.

(07:17):
We're describing what we think the universe is actually doing
when electrons fly through the universe. But you know, most
people who think about science, who are in philosophers naturally
imagine that that's what's happening, you know. They think that
what we see is real, that the electron is real,
not just something that we use to do calculations. If
you tell most people out there, like, look, we don't

(07:38):
know if an electron is actually real, or if the
theories that we use to describe them are just things
in our head. In your view, what's the sort of
easy way to understand or to approach the question of like,
what are we learning? Does it represent what's actually out
there or just things that we are imagining, you know,
without going full matrix sort of theory on you, what's
an easy way to access this question?

Speaker 2 (07:58):
In your view, the question is something like the extent
to which our theories should be taken as literally true
in their description of the world, and whether our evidence
for you know, taking our theories is literally true, whether
the success of our theories gives us good reason to

(08:19):
think that they are true in all their aspects, not
just like the stuff that they say about what we
can see, but the stuff that they say about what
we can't see. So some folks in philosophy think that
the empirical success of our theories, and they are like,
incredibly empirically successful, our best theories, gives us good reason

(08:43):
to just take them literally as true in all their aspects.
Other folks in philosophy think that the proper conclusion is
something more modest, like, you know, it isn't an anti
science stance or anything, but sugg just the empirical success
of our theories doesn't tell us that we should take
them as literally true both in describing stuff that's observable

(09:08):
and in describing stuff that's not observable. It just tells
you that theories are successful, and they were designed in
such a way to be successful. But are they uncovering
deep truths about the unobservable structure of the universe? Maybe not.

Speaker 1 (09:25):
Yeah, that's a great way to think about it, the
observable versus the unobservable. And I imagine people might be thinking, like,
what do you mean unobservable? We do experiments, we see stuff,
we know electrons are out there. What's a good way
to think about the unobservable sort of side of science.
One thing I sometimes imagine is like the fields themselves.
We talk about on this podcast a lot how space

(09:46):
is filled with fields, and sometimes those fields ripple to
make particles, or they ripple to make photons, et cetera.
But those fields aren't something we ever directly interact with. Right,
Is that an example of something that's unobservable but a
part of our theory.

Speaker 2 (09:59):
Yeah, that's right. So, like another good example comes from
the history of classical space time theories. So you can
think about Newton's old theory of gravitation. So, according to
Newton's old theory of gravitation, space time is flat. So
like the arenea in which events take place is flat,

(10:23):
it doesn't have interesting geometrical features. And according to this theory,
gravity is best described as a force, So it's a
field on your flat space time. And Newton's old law
of universal gravitation then dictates how particles massive bodies will
move around depending on what this force is at different
points in space time. It turns out that you can

(10:45):
do this theory. This was discovered by folks in the
early twentieth century. You can do Newton's theory, but on
a space time that's curved. So, in this alternative formulation
of Newtonian gravitation, space time is curved. Gravity manifests itself
and in terms of the curvature of space time rather

(11:06):
than force. And so now this raises an interesting question.
You can't observe you're not observing the structure of space time, right,
what you're observing is how stuff is moving around in
space time. And so now we have this question. It's
space time curved or is it flat? According to Newton's theory,
there are ways to do the theory according to which

(11:27):
it's curved, ways to do theory according to which it's flat.
They result in the same empirical prediction.

Speaker 1 (11:31):
But way, But how does the curved space Newton's theory
different from Einsteini in general relativity.

Speaker 2 (11:37):
It is the same basic concept as Einsteini in general relativity,
but it yields the same predictions as Newton's theory yields.
I see. So Einstein's theory and Newton's theory disagree about
like for example, the famous cases, they disagree about how
mercury would orbit the Sun. Right, this was the evidence

(11:58):
that led us to Einstein's theory was Newton's theory was
getting Mercury's orbit wrong. So the Newtonian gravitation theory set
on a curve space time, which folks will call newton
kartong theory, sometimes yields the same empirical predictions as Newton's theory,
not as Einstein's theory, but it agrees with Einstein's theory

(12:20):
on the status of the curvature of space time.

Speaker 1 (12:23):
I see, Okay, So you're saying that you could have
two theories. Newton's original theory of gravity is a force
and this other weird variation where gravity comes from the
curvature of space time, but it gives the same predictions
as Newton's original theory rather than the predictions of Einstein's theory.
And so then you're saying, this's an unobserved part of
the universe, the actual predictions of how things are going

(12:46):
to move. And then there's the sort of behind the
scenes mechanism. Is it a force or is it the
curvature of space time? What's sort of happening behind the
scenes to make that happen. So you're saying, there we
have like two theories with different descriptions for what's unobserved.

Speaker 2 (12:58):
That's right, There is different descriptions for what's un observed,
but the same exact description of observable stuff.

Speaker 1 (13:08):
That's really interesting and that touches on the topic I
wanted to dig into today, which is this underdeterminism, this
concept in philosophy that we might never actually reveal what's
happening in the unobserved section of the universe, that there
might be multiple ways to describe what's sort of happening
behind the scenes that give the same exact empirical predictions

(13:31):
that you could never distinguish in experiment, but be described
by different ideas. And you know, in the example you
just described, we know that both of those are actually wrong, right,
Newton's theory of forces in Newton's theory of kurt space
are both give the same predictions, but they're wrong. But
let's imagine some other scenario where we have a theory
of quantum fields and some independent group of scientists have

(13:51):
been working for a thousand years and they have a
theory of quantum shmields or something, and they're you know,
fundamentally different. They're not conceptually the same, but they give
the same prediction. So I think it's fascinating to think about.
You know, is it possible to have two really different
descriptions of what's happening in the universe that predict the
same thing that we would observe. Is that the sort

(14:12):
of the fundamental question of underdeterminism and philosophy.

Speaker 2 (14:15):
Yeah, that's exactly right, Like the underlying concept is simple
and will be familiar to all sorts of folks, Like
we find ourselves in situations just in our day to
day lives, where the body of evidence that we've gathered
doesn't help us adjudicate between two theories. So imagine like

(14:36):
you come home and I've been to the grocery store.
You're wondering what I bought at the grocery store. You
see the receipt sitting on the dining room table, and
you look at it, and all it says on the
receipt is that I spent eighteen bucks at the grocery store.
You know that apples they're like two dollars, oranges are
three dollars. And now you're faced with a kind of

(14:57):
underdetermination problem. You don't know how many apples I bought,
how many oranges I bought. The data gap doesn't help
me discriminate between the two theories. The data doesn't tell
you which one is correct, whether I bought six apples
or nine oranges or some combination. And that's exactly the

(15:18):
issue that the problem of underdetermination. That's exactly the issue
that the problem of underdetermination alleges we often face in
science where we have two theories and the data doesn't
help us decide which one is correct. The two theories

(15:40):
are compatible with precisely the same data, And so what
do we do in a case like that? What do
we believe right?

Speaker 1 (15:46):
And in that scenario it makes me feel uncomfortable that
I'm being so nosy about your grocery store purchases. But
you know, in the case of the universe, I don't
think the universe deserves any privacy. I think, you know,
we are entitled to peak behind the curtain as much
as we can and trying to figure out what's going
on and what the universe bought at the grocery store.
But what is the motivation for this? Is this just

(16:08):
like you know, philosophical meandering, Oh what if? What could
it be possible? Some sort of deep skepticism that maybe
will never figure things out, or do we have like
any sort of concrete examples of scenarios where we really
had two excellent theories that gave exactly the same predictions
but had different scenarios behind the scenes.

Speaker 2 (16:28):
Yeah, so let me give a couple of examples. The
example we were discussing earlier of Newton's standard theory of gravity,
in which gravity is a force versus this geometrized version
of theorem gravity. That's a famous historical case. Of course,
as you say, like, we know that neither of those

(16:50):
theories is correct. They make bad predictions, but the existence
of an example like this should give one pause. It
gives us read think that maybe the same thing can
happen with our current best theories. So there are other
examples from the history of physics. Another famous one is

(17:13):
the case of Hamiltonian and Lagrangian mechanics. Okay, suppose you
formulate your theory of classical mechanics. Newton's old theory how
stuff will move around in space by taking positions and
velocities is fundamental. Okay, So you think that we should
specify the energy conditions of a system by laying down

(17:37):
something that you call a Lagrangian So kind of a
function on possible configurations of positions and velocities that dictates
how active or lively these configurations are. Hell me how
these properties of the system, the positions and velocities of

(17:58):
particles in your system, will evolve over time by laying
down a set of equations that you call the Euler
Legronge equations. Okay, so call your theory Lagrangian mechanics. On
the other hand, for whatever reason, don't like velocity, Okay,
I find momentum to be much more elegant. You know,
it's conserved, and so I formulate my theory by specifying

(18:22):
energy conditions to the system laying down a Hamiltonian. I
take positions in momentum as the fundamental properties that a
system has, and I can tell you how the system
will evolve over time by laying down a different set
of equations. So call these Hamilton's equations, and they take

(18:43):
in the Hamiltonian of the system. So the Hamiltonian describes
something like the total energy of the system, rather than
its activity or liveliness, just total energy. I call my
theory Hamiltonian mechanics. So it turns out that our theories
are empirically equivalent. So what do we mean by that?
We mean, there's no possible evidence that is compatible with

(19:07):
my theory but incompatible with yours, or vice versa.

Speaker 1 (19:11):
So first, like some set of balls or squirrels on
roller coasters, they always give the same predictions for what's
going to happen exactly.

Speaker 2 (19:18):
Your theory gives the same prediction as my theory. But
it seems like we've done things differently, right, You care
about positions and velocities. I care about positions and momentums.
The laws of nature, one might say, according to your theory,
or different than the ones according to my theory, Like
we laid down different equations when we were saying how

(19:42):
the stuff was going to behave And so one might
think that we have a choice to be made here
between these two theories. But the evidence won't make the
choice for us. So that's another example of a possible
case of under termination, where two theories compatible with the

(20:02):
same body of evidence, but it seems like we might
have a choice to make between them. In cases like this,
the standard conclusion to draw is kind of a skeptical thing.
The things about which these two theories disagree. We now
don't have answers to these questions unless we can decide

(20:24):
which one of the two theories is correct, and how
do we make that decision. The evidence isn't going to
make the decision for us, so we have to come
to a decision by some other means.

Speaker 1 (20:35):
So we have two different sets of equations that give
the same predictions for what happens to squirrels on roller
coasters or billiar tables and these kinds of things, and
you're saying that they really are different in some way,
And because they always give the same predictions, there's no
way to do this sort of famous scientific test of say, well,
let's do an experiment and figure out the different hypotheses

(20:55):
and the different predictions, because they're always the same. But
if they're always the same, how do we know that
they really are different theories? I mean, I can take
any arbitrary theory and say I'm going to multiply everything
by two, and then at the end I'm going to
divide by two, and superficially, the equations might look a
little bit different, but it gives exactly the same predictions.
But not for an interesting reason, not because I'm really
saying this different stuff happening behind the scenes of the universe.

(21:18):
So in the case of Hamiltonian versus Lagrangeen mechanics, or
really in the case of fields versus Shmields, or any comparison,
what does it mean to say that they're different? How
do we know that they're different? Or is there some
standard we apply to say, like, this theory is telling
a different story about what's happening in the universe than
that theory. How do we do that?

Speaker 2 (21:36):
Yeah, good question. So, like the suggestion is something like,
maybe in some of these worrying cases of underdetermination, you
don't actually have two genuine rivals. They're the same theory,
just presented to you in different guises, And so like
you don't have a real choice to make. It's the
choice between one theory and itself, just presented to you differently.

(22:00):
So maybe let me tell a story. The most famous
case of this happened in the early years of quantum mechanics. Okay,
so in brief, Heisenberg had his matrix mechanics, Schrodinger had
his wave mechanics, and these guys like did not vibe

(22:22):
on the base of it. Their theories seemed incompatible. So
they made the same predictions, but they used like radically
different mathematical apparatus. To do so. So Schrodinger at one
point equipped that Heisenberg's theory lacked visualizability, into which Heisenberg
shot back, I quote what Schrodinger writes about visualizability as crap.

(22:50):
We eventually, like, we realized that there's a way to
translate between the Heisenberg picture and the Schrodinger picture and
vice versa. And so there's a sense in which there
are two ways of doing quantum mechanics were not that different.
After all. The disputes between the two theories, like the
stuff that they disagreed about is just verbal. It was

(23:13):
like we were having a verbal dispute. It was the
same theory presenting in two different ways. And so we
don't have a trouble in case of underdetermination. There a
trouble in case of underdetermination was avoided because you didn't
have a genuine decision to be made between these two
Between these two theories, the two theories were actually one.

Speaker 1 (23:34):
So it turns out that they were really just the
same theory dressed in different clothes, right.

Speaker 2 (23:38):
Yeah, yeah, And so this is something that one might
be tempted to say in the case of Hamiltonian and
lagrondgy mechanics too, you know, you pick position and velocity
to do your theory. I pick position and momentum to
do my theory. But we can translate back and forth
between these two descriptions. It's not like when you talk

(23:59):
about position in the law, I can't do that in
my theory. I would do it, you know, a little differently.
I'd use a little bit of different language to do so.
But am I saying a genuinely different thing about the
world when I decide to say it in position and
momentum language versus when you say it in position and

(24:19):
philosophy language. So in general, this is a kind of
response one can give to underdetermination worries. It's like, look
in a lot of the problematic cases of underdetermination that
we see, so cases where you have two theories that
can't be discriminated between on the basis of the evidence
we've gathered. In a lot of these cases, you don't

(24:41):
actually have genuine rivals. The two theories can't be discriminated
between on the basis of the evidence we've gathered, but
they're the same theory.

Speaker 1 (24:48):
Yeah, and so what about the case of later sort
of quantum mechanical interpretations. You know, we have various descriptions
of what's going on with particles. Is the way function
call colapsing, or is the universe splitting into multiple universes
or is the collapse relative? You know, like in relational
quantum mechanics. Are those examples of true underdetermination where we

(25:11):
have really different stories, genuinely different accounts of the physical world,
but that make effectively the same predictions for what we
could see in experiments, like we can't distinguish between the
many world hypothesis or the Copenhagen interpretation using experiments. Is
that an example?

Speaker 2 (25:27):
That is an example where it becomes harder for one
to say that the theories are genuinely the same, and
yet they have the same experiments supporting both of them.
So some folks at this point will say that there's
another natural way to respond to problems of underdetermination. So

(25:48):
the empirical evidence doesn't help us decide which theory is correct,
But we have some stuff that we can use to
make these decisions not using empirical evidence, and we do
this all the time. We appear to other virtues that
a theory might exhibit.

Speaker 1 (26:02):
Like simplicity or exact money.

Speaker 2 (26:05):
Okay, yeah, simplicity, fruitfulness, I don't know something like this,
And scientists do this all the time. We all do this.

Speaker 1 (26:15):
Well, I was having a conversation with a theoretical physicist
who's quite well known in particle physics, but I won't
name him, and I describe this problem to him, and
he said, well, look, if the theories predict the same
things but are different, then they're different only in the metaphysics, right,
only in the irrelevant details, which I think is another
way to say, like, well, maybe who cares? Right, Like,
we have two different descriptions of the universe, but they

(26:37):
give the same predictions. What does it matter? And to
me this is sort of shocking coming from a theorist
who you know, I think their job essentially is to
uncover the mechanisms of the universe, not just to produce
calculational tools that will get us to the next step,
but like reveal the nature of reality man. And so
to hear somebody be like, well, maybe it doesn't matter

(26:57):
if it's fields or shmields as lung as the numbers
are right.

Speaker 2 (27:01):
How does that strike you, as sort of.

Speaker 1 (27:03):
A philosopher of science, do you think that that sort
of instrumental approach is a solution to this problem, or
are we just avoiding the question.

Speaker 2 (27:10):
It's not an attitude that's uncommon among philosophers. Also, I
do think it's kind of avoiding sidestepping the question rather
than taking it on head on. One thing that I
should say, though, is notice that this kind of attitude.
So we have two theories, the evidence doesn't distinguish between them,

(27:32):
so they're the same in all important aspects. There's a
sense in which that attitude lands oneself in the same
place as taking the problem of underdetermination seriously ends oneself in.
The idea is, if you don't think that the theories
are telling you anything about the non empirical stuff, the

(27:54):
unobservable stuff, there's a sense in which you're not taking
your theories all that serious right, which is very close
to the place that the problem of underdetermination would lead
us in both cases, So, say you don't take the
metaphysics of your theory seriously, and there are some unanswered

(28:14):
questions like is space flat or is it curved? Suppose
you take the problem of underdetermination seriously, then you end
up with unanswered questions for a slightly different reason, but
in both cases you're ending with questions that are not
being answered by your physics.

Speaker 1 (28:31):
Okay, I have a lot more questions about how this
all works and how we can make sense of it,
but first let's take a quick break. All right, we're
back and we're talking to Professor Thomas Barrett about the

(28:53):
question of underdeterminism in philosophy of science, whether it's possible
to have multiple theories that describe the universe and make
exactly the same predictions but are in themselves fundamentally different. Well,
in almost every example, we've considered various versions of Newtonon theory.

Speaker 2 (29:11):
Et cetera.

Speaker 1 (29:12):
One can sometimes imagine that maybe in the future or
somebody will come up with an experiment to help us distinguish,
or somebody will be the next generation's John Bell will
come up with a super clever quantum experiment to help
us reveal multiple universes or something, even in the face
of like infinite data. Is the argument for underdetermination really
that it's impossible that, you know, even in a million years,

(29:33):
a billion years of super Einstein's could never come up
with a way to distinguish that. It might literally be impossible,
not just we haven't figured it out yet.

Speaker 2 (29:43):
Yeah, So think back. This is a good question think
back to the apples and oranges example. The natural thing
to say in that example is like, look, dude, you've
just looked at the receed on the table. You want
to know how many apples and oranges there are? Like,
go open up the pa entry, right, Like gather more data?

(30:03):
So problematic, like truly problematic cases of underdetermination, you know,
the ones that will keep you up at night. Our
cases where even gathering the more data won't help you
distinguish between the two theories. Just the two theories are
provably compatible with precisely the same collection of evidence. So

(30:27):
maybe one example, one further example will be helpful here.
So this comes from our recent work my good friend
and your colleague, get Irvine JB. Mancheck has done in
general relativity. So what Manchik proves is that in the

(30:47):
context of Einstein's theory of general relativity, we encounter a
very pernicious kind of underdetermination. So here's the gist. In
general relativity. Intuitively, like and signals can't travel faster than
the speed of light, you can only know at most

(31:08):
everything that goes on in your past light cone. So
all the evidence that's available to you at some time
will have arrived to you via some trajectory that's contained
in your past, like Cone. Right, like picture the stuff
that you see, Well, that's arriving to you via photons
that are reflecting off the surface of the thing. They're

(31:29):
all contained in your past like Cone. Your colleagues reporting
to you their evidence, well they came to you in
your past, like Cone. So Manchik proves now that given
this information, there's always more than one model of the
universe according to Einstein's theory that's compatible with the data

(31:49):
you could have possibly gathered.

Speaker 1 (31:51):
Meaning that there's always part of the universe sort of
shrouded in darkness because it could be just like so
far away given the age of the universe, that there
hasn't been time for light from it to arrive here,
And so there could be huge purple dragons beyond the
ende of the universe, or not huge purple dragons beyond
the edge of the universe.

Speaker 2 (32:07):
Yeah, exactly. I mean, they don't even have to be
that far from you. They just have to be not
in your backward like Cone. So the idea is, if
general relativity is true, if Einstein's theory is true, that
it itself implies that we can't know the global structure
of the universe because of the limitations that the theory

(32:29):
places on how evidence must work. It has to be
contained in your backward like cone, we have a case
where no matter how much data you gather, you're going
to have different possibilities for what the universe is. Like
that the data doesn't distinguish between well.

Speaker 1 (32:49):
It seems to me in the end to come down
to a form of skepticism to say, like, look, there
could always be some other theory out there that makes
the same predictions but has a different story about the universe,
and so is the conclusion then, I mean, what does
that mean about our understanding of the universe? As you say,
does that mean we shouldn't take too seriously the story

(33:11):
about what's happening When we write Fineman diagrams there are
a great little trick for making calculations. But does that
mean that we can't really believe that the universe is
doing its own finement calculations that fields are real because
there could be schmields instead. Does that mean we can
never really know what's happening out there in the universe.

Speaker 2 (33:30):
Yeah, So it all turns on how seriously you take
this possibility of alternative theories that have the same empirical
consequences as are current theories. So notice all we've done
here is I've given a bunch of examples of pairs

(33:51):
of theories where the evidence can't tell you which one
is rund. I haven't given you a general prison siezure
that you can use to construct from our best physical
theory arrival that agrees with it in all of its predictions,
but it's different. So that would be the gold standard,

(34:13):
right If we could do that, then I would be
very disturbed. We would have a mechanism for generating pairs
of theories where the evidence doesn't hell which one is right,
but they're different, and then we're kind of we're sunk.
I don't know how to answer our questions about the
unobservable anymore. We don't have that. All we have is

(34:34):
a bunch of examples. So what's the right conclusion to draws?
We do have a lot of examples. I mean, I've
talked to you as well. I don't know, depends on
how many you think is a lot. I think I've
talked about four.

Speaker 1 (34:44):
Let me just make sure I understand what you're saying
you're trying to be careful not to overstate your conclusions.
You're saying, we have some examples that suggest that it
might be possible to have two theories that explain the
universe equally well but have different stories, But we don't
have a general proof that that's all we possible. They
don't have like a procedure that says, if you have
a great theory, I can always use it to generate

(35:05):
an alternative theory that works just as well. So you're
saying there'd be a little bit of skepticism about our skepticism.

Speaker 2 (35:12):
Yeah, that's right, or we should be modest about our skepticism. Right.

Speaker 1 (35:16):
I was reading this article of by Philip Kitchener, and
he says, give us a rival explanation, and we'll consider
whether it is sufficiently serious to threaten our confidence, which
is basically like saying, maybe this is a problem in theory,
but right now it's not one that's facing us, so
maybe it won't.

Speaker 2 (35:33):
Yeah. I think that's right. So now philosophers have attempted
through the years to give kind of algorithmic procedures for
generating rival theories that are equally compatible with our data.
The kind of algorithmic procedures they give, though, result in
theories that most folks would not take seriously. So let

(35:55):
me give one example. So you have the theory of
the universe. I have a rival theory, like we'll call
it the when you turn your back theory. My theory
says that the universe behaves exactly like your theory says
it does when it's being observed, but when it's not
being observed, it behaves in some completely other, specific and

(36:16):
compatible Why our theories are equally compatible with the evidence.
My theory says exactly the same thing about how universe
will behave when it's being observed, as your theory does
by construction. But my theory is different from your theory. Okay,
So I mean, like take this over to the physics
department and tell a physicist. I mean, you can tell
me no one would take this theory seriously. Right, this

(36:36):
is ad hoc. It's totally gerrymandered, it's not interesting. It's
not a scientifically compelling theory. My theory that is. And
so it's like easy to dismiss this. It's not a
genuine case of underdetermination.

Speaker 1 (36:50):
Yeah, that's right, that's not something I would take seriously.
I mean, it's essentially equivalent in all the important bits. Right,
It's using the same calculational machinery to make predictions for
actual observations. Is just adding some bells and whistles to
the non observed side of things. It doesn't really seem
to me to count as a compellingly different explanation for

(37:11):
what's happening in the universe.

Speaker 2 (37:13):
Yeah, and going back to something you said earlier, like,
we have other means by which we can choose what
theory to believe. So our theories are equally compatible with
the evidence, but your theory is much simpler than mine, right,
Yours has a kind of elegance that mine lacks. So
we might think that that gives us better reason to

(37:34):
believe your theory than mine.

Speaker 1 (37:37):
And so is that the strongest argument against underdeterminism to say, like, look,
we don't have a perfect example, and we can't arbitrarily
generate one, so we don't really know that this is
a problem.

Speaker 2 (37:50):
Yeah, I would say there are two kinds of arguments
against the problem of under determination. So one is this
one that you just mentioned, Like how common are genuine
cases of underdetermination in order for this to be a
troubling problem. They had better be really common. The other
route is the thing that we were just talking about,

(38:10):
so appealing to what folks will call theoretical virtues that
are non empirical, so simplicity or fruitfulness or elegance, in
order to decide between rival theories that are equally compatible
with the evidence. So just because two theories are equally

(38:32):
compatible with the evidence, doesn't mean that we don't have
better reason to believe this one than we have reason
to believe this one. That opens a whole can of worms. Though, like,
these are philosophy questions. What are the non empirical reasons
for believing something? What are the reasons that are not
exhausted by just looking at the evidence for believing one
hypothesis over another. That's philosophy, right.

Speaker 1 (38:57):
Right, And you say that in a way that makes
it to like it's philosophy, it's not science. But I
wonder sometimes if we're too crisp about, you know, making
a delineation between those two things, because, as you say,
often we're using philosophy to make decisions in science, we
prefer this theory to that theory because it's simpler. Those
are choices influenced by our philosophy of science. We're doing

(39:18):
that kind of stuff all the time. The thing I
find fascinating is that most people in particle physics have
very strong opinions about philosophical questions, but they also often
think philosophical questions are a waste of time, you know,
like if I go around, sir and ask people like,
do you think the top quark or the Higgs boson
is real? It's there when we're not looking, or it's

(39:38):
just a tool in our calculations that are like, dude,
we discovered the Higgs boson. Here, we know it's real.
We found it, there's a Nobel prize for it. You're crazy.
You have to take them on a pretty long walk
to get them to the place where you're like, Okay,
that's true. It's not directly observed and therefore we don't
really know if it's there and there could be another explanation, etcetera, etc.
All right, I want to hear more about that, but

(39:59):
first we have to take another quick break. All right,
this is Daniel and we are doing an episode on
underdeterminism in science, wondering whether it's possible to explain the

(40:21):
universe in two different but equivalent ways. Well, my question
to you is, what do you think are the prospects
for making progress? Like is this something we can ever
really resolve? I mean, could it be that in a
million years we're doing science and we're still worrying about
this kind of thing, Like maybe somebody's going to come
up with another theory that explains what's going on. Or

(40:43):
do you think philosophers can like resolve questions like this,
that they can come up with a proof that two
theories that give the same predictions are fundamentally categorically the same,
or that you could develop the kind of algorithm you're
talking about earlier to generate an alternative theory that's realistic
from an existing theory. What do you think are the
prospects for making progress on this question?

Speaker 2 (41:03):
One thing I think is important to say is like,
in terms of progress, like setting aside under determination stuff
like science is always making progress. Right, taking the problem
of determination seriously doesn't imply that it isn't like progress
that's made when we moved from Newton's theory of gravity
to Einstein's, like, our theories get more and more predictively successful,

(41:23):
and that is definitely progress. So even if we still
take the problem of underdetermination seriously. Science is making a
kind of progress, So what about making progress on the
problem of underdetermination itself? I wish I knew. Note that
one of the routes that we were talking about in

(41:44):
responding to the problem just in fact, all of them
involve one doing like philosophy taking seriously the kinds of
questions that philosophers of science takes seriously. We have to
identify what the theoretical virtues are, so like what the
reasons are we have to believe one theory rather than another,

(42:08):
and then we have to argue that these help us
make good decisions, like in terms of which theory we
should believe. We also have to, as you were mentioning,
like come to some kind of agreement on win two
theories or saying the same thing, or when their genuine
rivals to one another. And these are hard philosophy questions.

(42:30):
So I mean, going back to something that you said earlier,
I think there's this famous Fineman quote. So he once
said that philosophy of sciences is useful to science or
to scientists, as ornithology is to birds. Oh man, I mean,
it's a great quote. Makes me chuckle. Every time, but
man like, look, birds are doing some ornithology here, right,

(42:53):
or or implicitly at least, or we have to in
order to make progress on this kind of stuff.

Speaker 1 (42:59):
Yeah, it's true, it's a great zinger, but I think
it's it's not necessarily very productive. Well, let me make
the question a little bit more sort of vivid and concrete.
Let's imagine a hypothetical scenario, say far in the future,
or maybe not that far in the future. Aliens arrive
and they are you know, scientific, and they use mathematics,
and they do physics, and they have also been pursuing

(43:22):
the project of revealing the fundamental nature of the universe,
matter and energy and particles in space and time and
all that stuff. And we get to sit down across
the table from them and compare notes. So here we
have like a completely independent scientific tradition, but you know,
undimently pointing at the same thing. So we can avoid
questions of like, what aliens do science? And could we
communicate with them? What do you think the chances are

(43:44):
that their theory is compatible with ours, that it's the
same theory, but you know, written with different kind of squiggles,
dressed in different clothing, or the chances that they really
have come up with a completely different mechanism to describe
the universe. What do you think the chances are in
that scenario that our two theories are compatible or incompatible.

Speaker 2 (44:04):
Yeah, I don't know, man, Your guess is as good
as mine. If I had to guess, I'd say it's
pretty unlikely that anyone else does science and exactly the
way we do it, you know, like so much of
the way that we do science is grounded in accidental
things about what humans are, like, you know, we're medium sized,

(44:27):
we travel pretty slowly. It's grounded and the kinds of
things that we care about what we want to do
with our science, how our particular perceptual apparatus works, like
how you know, our eyes work, stuff like that, and
the way that we do science is grounded in accidental
facts about history too. Given all this, it's hard for

(44:50):
me to imagine that anyone else would do it exactly
in the way that we do well.

Speaker 1 (44:56):
That's one of the things that makes me excited to
talk to aliens about science, because you know, if they're
doing science the same way we are, then hey, we
might learn some cool science. And if they're not, we
might learn some cool things about ourselves and the way
that we explore the universe and the way like being
human colors how we are seeing the universe, and it's
so hard to sort of get out of our own heads,
and that's like maybe one way to do it.

Speaker 2 (45:18):
So, yeah, when you hear from them, let me know.

Speaker 1 (45:23):
I think when they do arrive, we should send the
philosophers first before we send the physicists, you know, in
case they're not so friendly. All right, Well, thanks very
much for chatting with me about this really fascinating question
in philosophy and in science, and in philosophy and science.
I really enjoyed it. Thanks very much for your time.

Speaker 2 (45:38):
Yeah, thanks for having me on, Danielle. It was a pleasure.

Speaker 1 (45:48):
Thanks for listening, and remember that Daniel and Jorge Explain
the Universe is a production of iHeart Radio. For more
podcasts from iHeart Radio, visit the iHeartRadio app, Apple Podcasts,
or where where you listened to your favorite shows.

Speaker 2 (46:07):
H