0:00

Need to hire someone yesterday? When

0:02

it comes to hiring, indeed is

0:04

all you need. Stop struggling to

0:07

get noticed on other job sites.

0:09

Indeed sponsored jobs helps you stand

0:11

out and hire fast by jumping

0:14

to the top of the page

0:16

for relevant candidates. The results speak

0:18

for themselves. Sponsored jobs receive 45%

0:21

more applications than non-sponsored posts. With

0:23

no monthly subscriptions and no long-term

0:25

contracts, you only pay for results.

0:28

And speaking of results. and just

0:30

the last minute, 23 hires were

0:32

made of Indeed worldwide. There's no

0:35

need to wait any longer. Speed

0:37

up your hiring right now with

0:39

Indeed. And listeners of the show

0:42

will get a $75 sponsored job

0:44

credit to get your jobs more

0:47

visibility at indeed.com/Wonder U.S. Just go

0:49

to indeed.com/Wonder U.S. right now and

0:51

support our show by saying you

0:53

heard about Indeed on this podcast.

0:56

indeed.com/wondering US. Terms and conditions apply.

0:58

Hiring. Indeed is all you need.

1:00

No matter how you travel. It's

1:02

good to have a plan. Some

1:05

people plan every minute. No sleep,

1:07

bus, museum, another museum. While others

1:09

only plan to take it day

1:11

by day. When it comes to

1:13

your finances, a plan is just

1:16

as important. You can count on

1:18

credit karma to give you the

1:20

insights you need to understand your

1:22

options so you can take charge

1:25

of your financial path and find

1:27

your way to money. Explore how

1:29

you can make financial progress with

1:31

confidence today. Into it credit karma.

1:34

Karma you can count on. Hello

1:36

everyone and welcome to the Mindscape

1:38

podcast. I'm your host Sean Carroll.

1:40

I'm a working theoretical cosmologist, among

1:43

other job descriptions. So recently there's

1:45

been some news in cosmology that

1:47

may or may not turn out

1:49

to be a big deal. This

1:52

is often how it is in

1:54

science, right? You get a result,

1:56

but of course, by the very

1:58

nature of having gotten a new

2:01

result, it's a hard result to

2:03

get. Otherwise... would have gotten it

2:05

earlier. So the first indications that

2:07

something interesting might be happening are

2:10

typically faint. And you know, you're

2:12

not sure whether they're on the

2:14

right track or not. But there's

2:16

a couple of different things that

2:18

have indicated that perhaps there are

2:21

kinks in the armor of the

2:23

standard cosmological model, the so-called Lambda

2:25

for cosmological constant, CDM for cold

2:27

dark matter. Not something that throws

2:30

away the whole big bang. scenario

2:32

or anything like that, but specific

2:34

details might need to be tweaked.

2:36

This is something that I could

2:39

have done a solo episode about,

2:41

but the data and exactly what

2:43

the data are telling us really,

2:45

really matter here, so I thought

2:48

it would be better to have

2:50

a true expert on the podcast.

2:52

So we're happy to welcome Mark

2:54

Emmy Kowski, who is my colleague

2:57

at Johns Hopkins, and someone I've

2:59

known for a long time. We've

3:01

written papers together, including suggesting the

3:03

idea of dark electromagnetism. in addition

3:06

to dark matter out there in

3:08

the universe. We don't talk about

3:10

that in this podcast. Instead, we're

3:12

talking about these accumulating possible anomalies

3:15

in cosmology. Most recently, there's a

3:17

survey called the Dark Energy Spectroscopic

3:19

Instrument, D-E-S-I, that has suggested that

3:21

perhaps the density of dark energy

3:23

is changing with time. which is

3:26

not what you would expect if

3:28

it was just a cosmetrical constant.

3:30

If it were a dynamical field,

3:32

you might expect something like that.

3:35

And there was a hint a

3:37

year ago that that was true.

3:39

Very recently, the hint is become

3:41

stronger. And there is another instrument

3:44

called the Dark Energy Survey. DES,

3:46

as opposed to DESI for the

3:48

Dark Energy Spectroscopic Instrument, that has

3:50

less firm results but also pointing

3:53

in the same direction that dark

3:55

energy might be evolving with time.

3:57

These are both amazing surveys. Interestingly,

3:59

they both look at galaxies, right,

4:02

out there in the universe, and

4:04

they look at the distribution of

4:06

galaxies and how they're evolving with

4:08

time and things like that. They're

4:11

both ground-based cameras that replaced previous

4:13

cameras. The Dark Energy Spectroscopic Instrument,

4:15

Daisy, replaced a camera at Kit

4:17

Peak in Arizona, and the Dark

4:20

Energy Survey replaced a camera in

4:22

Chile, the Victor Blanco telescope. These

4:24

hints that dark energy might be

4:26

changing with time are still tentative.

4:28

It's not completely clear yet, and

4:31

indeed at face value, it would

4:33

be remarkable if they were really

4:35

true because of the specific way

4:37

in which the dark energy is

4:40

evolving with time. So we're going

4:42

to get into that. But of

4:44

course I have to take advantage

4:46

of this to also talk about

4:49

other cosmological anomalies. The Hubble tension,

4:51

which we talked about with Adam

4:53

Reese some time ago, Mark turns

4:55

out to be one of the

4:58

world's experts in thinking about models

5:00

to explain the Hubble tension. Mark

5:02

was in on the ground floor

5:04

in thinking about the cosmic microwave

5:07

background as a cosmological probe and

5:09

also was the author of some

5:11

interesting ideas about what dark energy

5:13

could be back in the day.

5:16

So he's really the best person

5:18

to talk to talk to about

5:20

what the microwave background tells us,

5:22

what these galaxy surveys tell us,

5:25

and what the theoretical implications are

5:27

of all this stuff. I would

5:29

say that, hmm, right now I'm

5:31

still on the fence about whether

5:33

there really truly is something dramatic

5:36

going on, but it's absolutely a

5:38

legitimate possibility. Sadly, we're still going

5:40

to have to wait for even

5:42

better data to come in. That's

5:45

how science goes sometimes, but if

5:47

you listen to this episode, you'll

5:49

be well prepared. to understand what's

5:51

happening when that data does come

5:54

in. So let's go. Mark

6:00

Aminkowski, welcome to the Vinescape podcast.

6:02

Hello, pleasure to be here. Nice

6:04

to be talking to you on

6:07

this beautiful Wednesday morning. I know,

6:09

you're back in sunny Baltimore. I'm

6:11

here in Santa Fe, but yeah,

6:13

it's a reasonably nice day-to-day, a

6:15

little cooler than yesterday, but probably

6:17

more oxygen, right? Santa Fe is

6:20

at very high altitude. Yes, gets

6:22

me. So we... are here because

6:24

there's been a couple of, more

6:26

than a couple, of anomalies, challenges,

6:28

puzzles for everyone to call them,

6:31

with respect to the standard cosmological

6:33

model, which is nowadays known as

6:35

lambda CDM. So we're going to

6:37

talk about that, but let's first

6:39

explain what is the standard cosmological

6:41

model, and why do we believe

6:44

it? Give us a medium-sized intro

6:46

to where we are before we

6:48

have any anomalies. Okay, medium-sized intro

6:50

to where we are before we

6:52

have any anomalies. So we live

6:54

in a universe that we have

6:57

been observing for centuries, but I

6:59

would say over the past hundred

7:01

years in particular, our understanding of

7:03

the universe, which is everything that

7:05

we know as a given, it's

7:08

one physical system, has evolved tremendously.

7:10

And it sort of started. Yeah,

7:12

just under 100 years ago really

7:14

with Hubble's discovery that the universe

7:16

was expanding. So, you know, everybody

7:18

knows that the Earth spins around

7:21

the Sun and the Sun is

7:23

the center of the solar system.

7:25

Most people know that the Sun

7:27

is one of about 10 million

7:29

stars in our galaxy, the Milky

7:32

Way, and the Sun spins around

7:34

the center of the Milky Way

7:36

for the same reason that the

7:38

Earth spins around the Sun. So

7:40

I think he said 10 million?

7:42

Ten billion, sorry. Get to know

7:45

you're paying good time. Let's get

7:47

them up the galaxy, yeah, and

7:49

billion, sorry, ten billion stars. And

7:51

so the sun... It spins around

7:53

the center of the Milky Way

7:55

for the same reason the Earth

7:58

spins around the sun, and that's

8:00

because all of the stars in

8:02

the Milky Way generate a very

8:04

strong, gravitating field. And you might

8:06

then wonder whether our galaxy is

8:09

part of some larger structure, you

8:11

know, whether our galaxy is one

8:13

of 10 billion galaxies that spin

8:15

around each other, but it turns

8:17

out that the hierarchy ends there.

8:19

And our galaxy, it turns out,

8:22

is one of, you know, tens

8:24

of billions of billions of galaxy.

8:26

that are more or less the

8:28

same that we know about. But

8:30

the galaxies don't spin around from

8:32

each other. It turns out that

8:35

every galaxy is moving away from

8:37

every other galaxy, and this is

8:39

what Hubble discovered almost 100 years

8:41

ago, and the relative, the speed

8:43

at which any two galaxies are

8:46

moving away from each other is

8:48

proportional to their distance. And so

8:50

the interpretation of this is that

8:52

the entire universe is expanding. this

8:54

was discovered by Hubble, and it

8:56

turned out that it was kind

8:59

of convenient because Einstein had discovered

9:01

General Relativity 12 years before that,

9:03

and, you know, several people who

9:05

were studying General Relativity realized that

9:07

equations of General Relativity allowed for

9:10

such a universe that was filled

9:12

with a bunch of stuff where

9:14

everything was expanding. Everything was moving

9:16

away from everything else. So that

9:18

was sort of the birth of

9:20

the standard cosmological model. And since

9:23

then, we've discovered a bunch of

9:25

other things. Perhaps the next big

9:27

breakthrough was sort of in the

9:29

mid-60s. There was a discovery of

9:31

something that we now call the

9:33

cosmic microwave background. Basically, the idea

9:36

is that if everything is moving

9:38

away from everything else today, If

9:40

we were to make a movie

9:42

of that expansion, then run it

9:44

backwards, at some earlier time, everything

9:47

in the universe would be on

9:49

top of everything else. So although

9:51

the universe is a fairly low

9:53

density placed now, if everything's moving

9:55

away from everything else, it's some

9:57

time in the past, which we

10:00

call it the big bang, the

10:02

density. of the universe would have

10:04

been very high. Anybody who puts

10:06

lots of air and tires and

10:08

drives them around knows that when

10:10

densities get high, the pressures get

10:13

high, the temperatures get high. So

10:15

the early universe, we have good

10:17

reason to believe, was very hot.

10:19

And you know, if you look

10:21

at a fireplace where there was

10:24

a fire that is now out,

10:26

the embers still glow for some

10:28

amount of time afterwards, even though

10:30

there's no fire, you can still

10:32

see residual heat. And in 1965

10:34

we discovered this residual heat, the

10:37

cosmic microwave background. So it turns

10:39

out that we discovered another relic

10:41

from this big bang that consistent

10:43

with this picture of an expanding

10:45

universe that Hubble sort of gave

10:48

us 100 years ago. Is this

10:50

good so far? This is great.

10:52

Yeah, I love it. Okay. Just

10:54

checking. So. And then... So that

10:56

was 1965, so that was 60

10:58

years ago, and since then we've

11:01

learned even more about our universe.

11:03

So we've been able to study

11:05

the distribution of galaxies in the

11:07

universe, and we find that the

11:09

universe on the very largest scales

11:11

is very, very smooth. So it's

11:14

like a pond on a clear

11:16

day, on a calm day. But

11:18

if you look very carefully, there

11:20

are some fluctuations. There are some

11:22

small amplitude ripples, as if there

11:25

was some... light wind. We've also

11:27

been able to look at this

11:29

cosmic microwave background very very precisely,

11:31

very carefully, and we've been able

11:33

to see that the temperature of

11:35

this, you know, glow, this afterglove

11:38

of the Big Bang, is not

11:40

precisely the same everywhere. It's pretty

11:42

close. You know, the temperature is

11:44

the same to one part in

11:46

100,000, but if you actually look

11:48

really, really, really carefully that are

11:51

small, there are small fluctuation. And

11:53

we believe, have very good reason

11:55

to believe that these small fluctuations

11:57

that we see in the cosmic

11:59

microwave background were then the seeds

12:02

for the... larger amplitude fluctuations to

12:04

see in the galaxy distribution of

12:06

Earth today. We believe that those

12:08

small fluctuations were amplified by gravitational,

12:10

you know, gravitational forces. So we

12:12

have all these very, very detailed

12:15

measurements of the cosmic microwave background,

12:17

of the distribution of galaxies, and

12:19

we have a model that allows

12:21

us to relate distribution of galaxies

12:23

in the universe today to the

12:26

distribution of the cosmic microwave background

12:28

that we see the afterglow from

12:30

the big bank. And in order

12:32

for our model to account for

12:34

the features that we see both

12:36

in the cosmic microwave background and

12:39

in galaxies, we need to have,

12:41

we need in these models, in

12:43

addition to the ordinary stuff that

12:45

you and I and everything the

12:47

solar system are made of, which

12:49

we call barionic matter, which jargon

12:52

for ordinary atomic stuff. In addition

12:54

to the barions, we also know

12:56

that there has to be a

12:58

lot of dark matter about five

13:00

times as much mass in dark

13:03

matter as in bariums. You don't

13:05

know what dark matter is, but

13:07

the models require that, you know,

13:09

the dark matter is required in

13:11

order for the models to work.

13:13

And then there's also something called

13:16

the cosmological constant. that was inferred

13:18

in the late 1990s, but we

13:20

now also understand from the models

13:22

that we have for these fluctuations

13:24

that it has to be there.

13:26

And then the cosmological constant is

13:29

something we don't really know what

13:31

it is, but in some sense,

13:33

it's some energy density that pervades

13:35

all of space. So we have

13:37

this great model, explains the origin

13:40

of the universe, why it's the

13:42

expansion of the universe. We have

13:44

some ideas about why it's expanding,

13:46

although those are not fully formed

13:48

yet, I would say. You mean

13:50

what started it in some sense?

13:53

Yeah, what set it in motion?

13:55

The good news for you is

13:57

that I have a future upcoming

13:59

podcast about what happened near the

14:01

Big Bang, so you don't have

14:04

to worry about that. Oh really,

14:06

near the Big Bang. What about

14:08

before the Big Bang? Oh yeah,

14:10

that's going to be there. Yeah,

14:12

that should be. fun. Okay, so

14:14

we have this great model that

14:17

explains all these this wealth of

14:19

observations we have with the galaxy

14:21

distribution. This is, you know, millions

14:23

and millions of galaxies that we've

14:25

been able to map. And the

14:27

temperatures of the cosmic microwave background,

14:30

we've been able to measure it,

14:32

you know, about a million different

14:34

points in the sky. So there's

14:36

a lot of data. It's not

14:38

just a hand-waavy, squiggly approximate model.

14:41

It's not like, you know, about...

14:43

3,000 miles from New York to

14:45

Los Angeles. It's, you know, 3,118,

14:47

632. And that's, you know, it's

14:49

a really good model. Right. And

14:51

we're really proud of ourselves. I

14:54

think you should be. Let me

14:56

pause though for a second because

14:58

something sneaked in there that I

15:00

think is really interesting. A lot

15:02

of people, I'm sure that you

15:04

get emails from people who have

15:07

explained away dark matter without being

15:09

professional scientists, etc. And of course,

15:11

they always concentrate on the rotation

15:13

curves of spiral galaxies. So this

15:15

idea that the amount, the rate

15:18

at which stars and gas are

15:20

rotating around the centers of spirals

15:22

depends on how much mass there

15:24

is, etc., etc. Ordinarily, Vera Rubin

15:26

and her collaborators proved this, we

15:28

attribute that to dark matter, but

15:31

it could be something else. But

15:33

you didn't even mention spiral galaxies.

15:35

You went right to the microwave

15:37

background. Yeah, that's a good point.

15:39

So I think I did that

15:42

because I was trying to give

15:44

you a capsule summary of a...

15:46

No, I like it. Yeah. The

15:48

model for the universe, but yes.

15:50

So the measurements of the cosmic

15:52

microwave background and... large-scale distribution galaxies

15:55

that I told you about that

15:57

implied the existence required the existence

15:59

of dark matter. Those happened about

16:01

25, started happening about 25 years

16:03

old, but you are correct that

16:05

even 20 years before that, you

16:08

know, around 1970, Vera Rubin and

16:10

her collaborators and a few other

16:12

people started to realize that most

16:14

of the matter in the galaxy

16:16

has to be dark. And so

16:19

we actually had reason to believe,

16:21

you know, we had good reasons

16:23

to believe that there would be

16:25

dark matter in the universe before

16:27

these large scale structure cosmic microwave

16:29

backgrounds that I told you about.

16:32

So in some sense it wasn't

16:34

a surprise when that happened, but

16:36

it was a confirmation and it

16:38

was, you know, gave us much

16:40

more confident that what we were,

16:42

that the anomalies that we were

16:45

seeing with galactic rotation curves were

16:47

actually real and due to some

16:49

new form of. Starting therapy can

16:51

seem like a big investment, emotionally

16:53

as well as financially, but the

16:56

state of your mind is just

16:58

as important as your physical health.

17:00

Let's talk numbers. Traditional in-person therapy

17:02

can cost anywhere from $100 to

17:04

$250 per session, but with better

17:06

help, you can save on average

17:09

up to 50% per session. With

17:11

better help you pay a flat

17:13

fee for weekly sessions saving you

17:15

big on cost and time. Therapy

17:17

should be accessible, not like a

17:20

luxury. With online therapy, you get

17:22

quality care at a price that

17:24

makes sense and can help you

17:26

with anything from anxiety to everyday

17:28

stress. With over 30,000 therapists, Better

17:30

Help is the world's largest online

17:33

therapy platform, having served over 5

17:35

million people globally. And it's convenient.

17:37

You can join a session with

17:39

a click of a button helping

17:41

you fit therapy into your busy

17:43

life. Plus, switch therapists at any

17:46

time. Your well-being is worth it.

17:48

Visit betterhelp.com/Mindscape today to get 10%

17:50

off your first month. That's better

17:52

help help.com/Mindscape. But the reason I

17:54

like to emphasize it is because

17:57

it does kind of highlight a

17:59

difference in how the professionals think

18:01

about this than how we perhaps

18:03

talk about it to the broader

18:05

public. We, you know, we tend

18:07

to be historically quasi accurate and

18:10

we want to give the early

18:12

people credit so we talk about

18:14

spiral galaxies, but the real reason

18:16

we are confident that there's something

18:18

like dark matter is much more

18:20

something like some combination of the...

18:23

microwave background radiation, large scale structure,

18:25

things like that. And so accounting

18:27

for spiral galaxies. doesn't actually get

18:29

you out of the need for

18:31

dark man? Accounting for spiral galaxy

18:34

does not get you out of

18:36

the need. Dark. Yes, yes, that's

18:38

right. If that's right. That's right.

18:40

That's trying to parse what she

18:42

said. If there were indeed, if

18:44

somebody had some other explanation for

18:47

the galactic rotation curve. that did

18:49

not involve dark matter. We would

18:51

still have reason to believe that

18:53

dark matter exists as of observations

18:55

of cosmic microwave background and galaxy

18:58

distance. Right. Yeah. Sorry to hectare

19:00

you on that, but it is

19:02

the internet that we're talking to

19:04

here and there are people out

19:06

there who have ideas. And we

19:08

love them and we support their

19:11

efforts, but we want to be

19:13

clear about why we believe these

19:15

things. Yeah, it's actually, I mean...

19:17

It's a good point that you

19:19

make and I think it's something

19:21

that we are becoming we've always

19:24

known but appreciate more in cosmology

19:26

with time and that is that

19:28

you know when we do cosmology

19:30

it's sort of like archaeology or

19:32

you know physical anthropology or paleontology.

19:35

Yes, paleontology. Missing the word there.

19:37

You know, with paleontology, what you

19:39

do is you find bones somewhere

19:41

and they have these funny looking

19:43

shapes, but you look at them

19:45

and it's sort of like a

19:48

puzzle and you sort of try

19:50

to put the pieces of the

19:52

puzzle together consistent with what you

19:54

know about, you know, bones of

19:56

animals that exist. So it's a

19:58

puzzle. But it's also informed by,

20:01

you know, your solution to that

20:03

puzzle is performed by other solutions

20:05

to similar puzzles you have. And

20:07

we do the same thing in

20:09

cosmology. It's very similar. It's not

20:12

an experimental science. We don't, like

20:14

in paleontology, you don't build a

20:16

dinosaur. Although some people are trying.

20:18

Yeah. You don't build the dinosaur.

20:20

You know, we can't alter the

20:22

system. We just have observations. There

20:25

are things that we find with...

20:27

telescopes. And so we try to

20:29

construct a model that's consistent. with

20:31

the observations and consistent with what

20:33

we know about the laws of

20:36

physics. And so, you know, if

20:38

we have a model for galactic

20:40

rotations that involves something other than

20:42

dark matter, that's a perfectly legitimate

20:44

thing to try. But then you

20:46

have to ask, is that is

20:49

that solution going to be consistent

20:51

with other things that I, right?

20:53

And now with cosmology, you know,

20:55

we try to make as many

20:57

different observations we can. try to

20:59

study as many different systems as

21:02

we can in detail. And, you

21:04

know, in some cases, there are

21:06

things we can try in the

21:08

laboratory, but basically in order to

21:10

actually have confidence and, you know,

21:13

conclusions that we make, in order

21:15

to, you know, increase our conference,

21:17

we want to have different measurement

21:19

and different observation from different systems

21:21

and different types of observational techniques,

21:23

that then all match and give

21:26

you. Speaking of which, this dark

21:28

energy business, this cosmological constant business,

21:30

where did we figure out that?

21:32

So the cosmological constant, I mean

21:34

the story is that Einstein had

21:36

this, you know, realize that there

21:39

must, there might be a cosmological

21:41

constant, the Einstein equations, and called

21:43

the Viscus Blunder, whether that's true

21:45

or not, I don't know. I

21:47

actually saw the notebook, the page,

21:50

you know, Diana Buchwald, the Einstein

21:52

papers problem. So Diana once showed

21:54

me the actual like notebook pages,

21:56

pages, pages in Einstein's notebook, where

21:58

he was doing the calculation to

22:00

let him to think of the

22:03

cosmological constant. And it was kind

22:05

of interesting, what she told me,

22:07

it's, um, she said it's the

22:09

only case that they have in

22:11

all of his papers, the only

22:14

example in all of his papers

22:16

where he was actually doing a

22:18

numerical calculation. Flugging him in his

22:20

papers. Yeah, he actually like had

22:22

a graph in those graph in

22:24

those graph paper. He was like

22:27

trying to calculate the area under

22:29

the curve. Oh, wow. He did

22:31

it by counting the boxes. Anyway,

22:33

you know, the cosmological concept sort

22:35

of existed as a possible theoretical

22:37

addition to the basic theory of

22:40

general relativity for over 100 years.

22:42

But, you know, the observational evidence

22:44

that that thing actually exists came

22:46

about in the late 1990s. And

22:48

you can look in the literature

22:51

even before the ninth. late 1990s,

22:53

people were sort of speculating that

22:55

various cosmological observations were better fit

22:57

with a non-zero cosmological constant, but

22:59

you know, the real smoking gun

23:01

was measurements made by two independent

23:04

groups, the supernova cosmology project and

23:06

the high Z supernova team. Yes,

23:08

the high Z supernova team. They

23:10

sound like the same thing. Yeah.

23:12

And so I told, you know,

23:14

we talked earlier about how... every

23:17

galaxy in the universe is flying

23:19

apart from every other galaxy. And

23:21

if you think about, you know,

23:23

a ball that I throw in

23:25

the air, if I throw a

23:28

ball in the air, it goes

23:30

up, but then experiences the gravitational

23:32

attraction to the Earth. And so

23:34

even though I throw it up

23:36

initially with some large velocity, the

23:38

velocities flows eventually goes to zero,

23:41

becomes negative, and then it falls

23:43

back down. Now, if I had

23:45

a really, really good arm, and

23:47

I could throw the baseball at

23:49

a velocity bigger than 11 kilometers

23:52

per second, I don't know what

23:54

that is in miles, if I

23:56

could throw a ball with a

23:58

velocity greater than 11 kilometers per

24:00

second, it would actually escape the

24:02

gravitational field of the earth. It

24:05

would actually, you know, instead of,

24:07

you know, going up and then

24:09

flying back, I would actually... that

24:11

ball actually fly away from the

24:13

earth and continue flying away from

24:15

the earth forever. But since gravity

24:18

is a long-range interaction, a long-range

24:20

attractive interaction, even though that baseball

24:22

is flying away from the earth

24:24

and would continue to fly away

24:26

from earth, you know. forever, the

24:29

speed at which it does so

24:31

would continually decrease. So, you know,

24:33

ordinary gravity, ordinary Newton's gravity, suggests

24:35

that, you know, if two galaxies

24:37

are flying apart from each other,

24:39

the relative speed that they fly

24:42

apart from each other should be

24:44

decreasing with time. And that is

24:46

in fact what was in the

24:48

standard cosmological model. based on Einstein's

24:50

general relativity with no cosmological constant

24:52

until the late 1990s. And then

24:55

what happened is that the High

24:57

Z Supernova Supernova Team and the

24:59

Supernova Cosmology Project independently actually measured

25:01

how fast galaxies were moving away

25:03

from each other is actually increasing

25:06

with time rather than decrease. And

25:08

this was science magazines breakthrough of

25:10

the year. 1998, it was completely

25:12

and utterly shocking to everybody in

25:14

physics. We knew that general relativity,

25:16

you know, could allow for the

25:19

possibility of a non-zero cosmological constant,

25:21

but everybody just assumed that it

25:23

would be zero, because the actual

25:25

value is something like, oh, 0.00,

25:27

0.100 with 120 zeros, 1. The

25:30

actual value is extremely, extremely, extremely

25:32

small. And physicists don't like extremely

25:34

small, or if we like one,

25:36

we like pie, you like 2.3,

25:38

we don't like extremely small or

25:40

extremely large, no. So everybody was.

25:43

Very, very shocked. I remember being

25:45

very, very skeptical. People tried to

25:47

explain it away. They tried to

25:49

suggest that maybe the supernovae themselves

25:51

were evolving with time. They speculated

25:53

that maybe light was being absorbed

25:56

by the more distant supernov, thus

25:58

making them look fainter. And, you

26:00

know, the people in both projects

26:02

did a really good job. you

26:04

know, checking all of these things

26:07

and dispelling, you know, all of

26:09

these possibilities, ruling out all these

26:11

possibilities. And then I became really,

26:13

really convinced when the cosmic microwave

26:15

background experiments came out in the

26:17

early 2000s. And, you know, from

26:20

a completely different type of measurement,

26:22

different type of observation, they also

26:24

inferred that there had to be

26:26

a non-zero value of the cosmological

26:28

constant. Did I answer your questioning?

26:30

You did, you did. And thus,

26:33

Lambda CDM, the Lambda for the

26:35

Cosmological Constant CDM for Cold Dark

26:37

Matter, that is the standard target

26:39

fiducial cosmological model. That is our

26:41

standard cosmological model. I don't like

26:44

the name. It's not the sexiest

26:46

name, but you know, maybe we'll

26:48

overturn it. So that's okay. We

26:50

can come up with a better

26:52

name. I guess the one other

26:54

piece of cosmological... measurement that I

26:57

wanted to get on the table

26:59

was the idea of a barion

27:01

acoustic oscillation. I think it's probably

27:03

the trickiest thing for the person

27:05

on the street to wrap their

27:08

brains around, but apparently very very

27:10

important to modern cosmology. Yeah, this

27:12

is the hardest thing to explain,

27:14

but I'll try. So we know

27:16

from our observations of the cosmic

27:18

microwave background that the early universe

27:21

was very, very smooth. So I

27:23

said it was sort of like

27:25

the surface of a pond on

27:27

a very calm day. But suppose

27:29

I threw a pebble into that

27:31

pond. There would be a splash,

27:34

but then there would be a

27:36

wave that propagates out from where

27:38

the pebble landed in the pond.

27:40

And so, you know, that wave

27:42

expands with time, and it's moving

27:45

at some velocity. And at early

27:47

times, you know, if I were

27:49

to take a snapshot just a

27:51

few seconds afterwards, the circle would

27:53

be small. That wave would be

27:55

small. And at later times, the

27:58

circle, that circle, wave would be

28:00

larger. Now I told you that

28:02

although the early universe was very

28:04

very smooth, it was not perfectly

28:07

smooth. And so there were sound waves

28:09

propagating the early universe. The early universe

28:11

consisted of this, you know, fluid, you

28:14

know, all the barions that make up

28:16

the galaxy and you and me, the

28:18

sun, all the other stars. All those

28:20

barions would have made up a fluid

28:23

in the early universe. If I have

28:25

a disturbance in the early universe, if

28:27

I were to throw a pebble into

28:30

the early universe, there will be a

28:32

wave that propagates out at the speed

28:34

of sound. Now, although we don't see

28:37

an individual such wave, what we do see,

28:39

you know, if I have a pebble in a

28:41

pond, then I throw it in the pond. There

28:43

will be that circle, which is

28:45

the, you know, the wave propagating

28:47

out, but there will also still

28:49

be some, you know, bubbling right

28:51

at the center. and so there

28:53

will actually be a correlation in the

28:55

surface height of the water at

28:57

the center and at the wave. Okay,

29:00

so the surface, you know, the surface

29:02

height far away from the wave

29:04

is zero, the surface height inside

29:06

the wave is pretty small, but

29:08

there will be, you know, an

29:11

increase in the surface height at

29:13

just this right distance. And so

29:15

when we look at the galaxy

29:17

distribute, we don't see any individual

29:19

wave, but we... can measure the

29:22

probability to find one galaxy at

29:24

some distance from some other galaxy.

29:26

And if you look at that

29:28

probability to find one galaxy at

29:30

some distance from some other galaxy,

29:33

that probability decreases as you go

29:35

to larger and larger radii, the

29:37

excess probability probability, the

29:39

excess probability, but then it turns

29:41

out that there's a bump, somewhere

29:43

around 100 megaparsex. And that bump

29:45

is actually, essentially a consequence of

29:48

these sound waves in the sound

29:50

waves in the early unit. Okay,

29:52

so it's roughly speaking metaphorically, God

29:54

threw pebbles at the smooth pond

29:56

of the early universe and ripples

29:59

went out. and there's going to

30:01

be sort of a natural correlation

30:03

length between the different galaxies that

30:05

we see today because of just

30:08

the time scales of which everything

30:10

happened. That is correct. Good. And

30:12

that's the barion acoustic oscillation, B-A-O.

30:14

Yep. It's pretty remarkable when you

30:17

see it in the data. Yeah.

30:19

I mean, I mean, this is

30:21

another example of the whole, you

30:23

know, everything hanging together. I mean,

30:25

you know, we had, you know.

30:28

the cosmic we have the expansion

30:30

of the universe, we have the

30:32

cosmic microwave background, we have these

30:34

cosmic microwave background fluctuations. And I

30:37

mean, in the, I remember when

30:39

I was a postdoc and assistant

30:41

professor in the mid-1990s, people sort

30:43

of understood that you should also

30:45

see a bump in the galaxy

30:48

distribute. But I remember thinking that

30:50

there's no way. we'd ever be

30:52

able to like see. It's an

30:54

interesting theoretical idea, but you know,

30:57

you need to, you know, measure

30:59

the positions of God knows how

31:01

many millions of galaxies to actually

31:03

ever see this. And even so,

31:06

you know, all kinds of complicated

31:08

things happen between the big thing

31:10

and now. But it turns out

31:12

that the model actually works and

31:14

you actually, you know, and we

31:17

actually do have surveys of millions

31:19

and millions of galaxies of galaxies

31:21

with very well-measured-measured-measured-measured-measuredured-pres-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me-me- And we see

31:23

this bump in the galaxy distribute,

31:26

and it's not at all subtle

31:28

now with current measurement, yes, you

31:30

know, really in your face. It's

31:32

a, I mean, we're actually, you

31:34

know, you look at the distribution

31:37

of galaxies on distant scales of,

31:39

you know, hundreds of millions of

31:41

light years, and you see the

31:43

imprint of this physics that we

31:46

have in our models to describe

31:48

the early universe. Well, and it's

31:50

important to emphasize that with both

31:52

the temperature fluctuations in the migrate

31:55

background and with the barion acoustic

31:57

oscillations, you are measuring in... principle,

31:59

all these parameters, like the density

32:01

of dark matter, the density of

32:03

the cause, module, constant, etc. But

32:06

you're not measuring them directly. You're

32:08

saying that I have a model

32:10

for everything at once, and I'm

32:12

going to make all sorts of

32:15

predictions, and those predictions will depend

32:17

on the parameters. I'm going to

32:19

measure some things, and then ask

32:21

which values of the parameters give

32:23

me the best thing. That is

32:26

correct. So on the one hand,

32:28

very, very impressive that the model,

32:30

the standard model, works so well.

32:32

There's a lot of moving parts,

32:35

right? If something doesn't start fitting,

32:37

then it's not going to be

32:39

perfectly obvious where to look. Yes,

32:41

that is correct. So another thing

32:44

I think that's been surprising about

32:46

our understanding, our evolution of the

32:48

understanding of the universe, is that

32:50

the universe has turned out for

32:52

reasons we probably don't fully understand

32:55

to be a much simpler system

32:57

than anyone might have surmised. So

32:59

you're right. measure the distribution of

33:01

bazillions of galaxies, we measure the

33:04

distribution of millions, you know, the

33:06

temperature of the cosmic microwave background

33:08

over millions of points on the

33:10

sky. It's a really complicated system

33:12

because galaxies have gas, they have

33:15

stars, they have, you know, interaction

33:17

between outflows from stars that are

33:19

supernovae that blow up and then,

33:21

you know, pollute the intergalactic medium

33:24

with heavier elements. There's gravity. There's

33:26

this cosmological constant, there's dark matter.

33:28

It turns out to be a

33:30

very, very complicated, seemingly very, very

33:33

complicated system, but it turns out

33:35

that it's much, when we look

33:37

at it carefully, the system as

33:39

a whole turns out to be

33:41

much simpler than anyone might have,

33:44

you know, our understanding of the

33:46

origin and evolution of the universe

33:48

is actually, I think, you know,

33:50

much more precisely parameterized it specified

33:53

by the model than is our

33:55

understanding of the solar system. Even

33:57

though we live in the solar

33:59

system, have visited parts of it,

34:02

and it's, you know, much simpler

34:04

physics in principle, you know, gravity

34:06

and Kepler's law. So it turns

34:08

out that you're right. We have

34:10

all this data, we have all

34:13

these parameters to turn. It seems

34:15

like it would be a really

34:17

complicated system. It would be hard

34:19

to have any confidence, you know,

34:22

have much confidence in any individual

34:24

parameter, given that you're trying to

34:26

simultaneously fit for these other parameters.

34:28

But it turns out that a

34:30

model with five parameters can account

34:33

for it all. And I mean,

34:35

that's been one of the things

34:37

that's been so, you know, so

34:39

surprising and impressive. You know, that

34:42

one model can explain the galaxy

34:44

distribution and the cosmic microwave background,

34:46

when that's why we were so

34:48

proud of ourselves. We should be

34:51

proud, but we should also be.

34:53

interested to see if there's anything

34:55

weird going on. I mean, in

34:57

terms of weird things that could

34:59

go on, there's like theorists' favorite

35:02

ideas, and then there's what the

35:04

experimenters actually come back and tell

35:06

us about. What is it, like,

35:08

very quickly, I think, what are

35:11

some of the main alternatives in

35:13

terms of perhaps the physics of

35:15

dark matter and dark energy that

35:17

we're trying to test when we

35:19

do the cosmic experiments? Oh, okay,

35:22

that's a good question. So. So

35:25

one thing that we do to

35:27

test the cosmological constant, so as

35:29

we said, the cosmological constant is

35:31

a very, very strange thing from

35:34

the point of view of fundamental,

35:36

our understanding of fundamental physics. And

35:38

so one thing that you can

35:40

wonder is whether the cosmological constant

35:43

is really constant. So it sort

35:45

of says that there's some mysterious

35:47

energy pervading all of space, but

35:49

it's, you can ask, is that

35:52

changing with time? as the universe

35:54

expands or is it really really

35:56

constant? And so if your cosmological

35:58

constant is not really... constant and

36:01

you know people have been using

36:03

the word dark energy in place

36:05

of cosmological constant because the cosmological

36:07

constant isn't constant then go down

36:10

before it. So you can ask

36:12

whether the dark energy density is

36:14

constant in time or evolving at

36:16

time. There's been a major effort

36:19

over the past 20... five years

36:21

to try to address this question,

36:23

try to figure out whether the

36:25

energy density is changing with time

36:28

or not. And that is sort

36:30

of done with the same types

36:32

of measurements that we use to

36:34

determine the expansion rate and, you

36:37

know, to determine dark matter density,

36:39

etc. We have models for how

36:41

the galaxy distribution, the cosmic microwave

36:43

background should look. those models have

36:46

incorporated into them as one ingredient

36:48

dark energy. You know, in the

36:50

simplest model is the dark energy,

36:52

the only parameter that we use

36:55

to describe the dark energy is

36:57

density, which we assume to be

36:59

constant, but you can also see

37:01

what happens if you have a

37:04

model where the dark energy density

37:06

evolves with time. And so we

37:08

have parameters now that we can

37:10

measure for fit from the model,

37:13

fit from the data with the

37:15

model. to figure out or see

37:17

if the dark energy density is

37:19

evolving with. Right. With dark matter,

37:22

it seems to be harder to

37:24

have sort of physically plausible modifications,

37:26

but people still do play around

37:28

with it. Yeah, that's actually in

37:31

some ways a bigger industry. You

37:33

know, we have no idea what

37:35

dark matter is. The models work

37:37

very well if we make the

37:40

simplest assumption that dark matter interacts

37:42

with itself and with everything else

37:44

only. gravitationally. So in other words,

37:46

dark matter particles don't scatter from

37:49

themselves, they don't scatter from the

37:51

ordinary stuff. But that's, you know,

37:53

an assumption. And again, you can

37:55

construct more complicated models where dark

37:58

matter has some type of interaction

38:00

with itself or some type of

38:02

interaction with ordinary matter. and then

38:04

you can describe those interactions of

38:07

terms of parameters that you can

38:09

then try to fit from the

38:11

data, you know, from the galaxy

38:13

distribution cosmic graph background. But with

38:16

dark matter, it's a little, we've

38:18

had a few more possibilities. You

38:20

know, the dark matter is not

38:22

only out there in the universe,

38:25

it's presumably also, you know, in

38:27

the Milky Way, and in the

38:29

solar system, and, you know, presumably,

38:31

passing through us here on Earth

38:34

every single day. So, you know,

38:36

one of the prevailing ideas for

38:38

dark matter is that it's a

38:40

elementary particle that has a mass

38:43

of roughly 100 times the proton.

38:45

And it turns out that the

38:47

dark matter density locally is roughly

38:49

half a proton mass per C.C.

38:52

And so what that means is

38:54

that, you know, every time you

38:56

buy a liter of milk at

38:58

the store, in addition to your,

39:01

you know, recommended daily allowance of

39:03

calcium and vitamin D, you are

39:05

also getting... you know, one dark

39:07

matter particle. I mean, if it's

39:10

axions, you're getting a lot of

39:12

dark matter particle. Yeah, yeah, it

39:14

could be. Yeah, I mean, the

39:16

question is, you know, are you

39:19

buying it by weight or by...

39:21

So, you know, as I said,

39:23

the canonical... idea for dark matter

39:25

is that it interacts with nothing

39:28

else except gravitationally, so you don't

39:30

have to worry about it if

39:32

it winds up in your milk.

39:34

But, you know, if it does

39:37

have some very weak interaction with

39:39

ordinary matter, then we can construct

39:41

laboratory detectors to try to, you

39:43

know, see the effects of interactions

39:46

of these very rare dark matter

39:48

particles with ordinary matter. So, you

39:50

know, that's a fairly big industry.

39:52

And so far we have seen

39:55

zero. But just emphasize there's plenty

39:57

of room for very very sensible,

39:59

viable, dark matter candidates that we

40:01

would not have seen yet. Yes,

40:04

that is correct. Well, I mean,

40:06

it's actually... it's interesting that I

40:08

think the people first started to

40:10

think about particle, elementary particle dark

40:13

matter, seriously about 40, 45 years

40:15

ago. So in the late 1980s

40:17

people started to get serious about

40:19

actually looking for these dark matter

40:22

particles in the lab. And so

40:24

we've been looking for dark matter

40:26

particles in the lab. for 40

40:28

years. And during that time, you

40:31

know, 40 years ago, we had

40:33

predictions or, you know, pervade, very,

40:35

very elegance, attractive predictions for what

40:37

the dark matter should be. And

40:40

many of those models have been

40:42

ruled up because we haven't seen

40:44

them. So in some sense, you

40:46

know, we don't know what dark

40:49

matter is, some sense a shot

40:51

in the dark, but we have

40:53

actually had, you know, over the

40:55

past few decades, a number of

40:58

really... intriguing and interesting and promising

41:00

theoretical models for dark matter and

41:02

you know it's interesting that we've

41:04

been able to rule those out

41:07

you know it's dark matter this

41:09

is an experimental science we're not

41:11

you know just casting up you

41:13

know flailing about completely the dark.

41:16

That's nice to hear but it

41:18

brings us smack into the fact

41:20

that we do have puzzles that

41:23

we need to deal with. I

41:25

guess chronologically the first puzzle that

41:27

I personally took seriously that is

41:29

still lingering is the Hubble tension.

41:32

In fact, we had our mutual

41:34

colleague Adam Reese on the podcast

41:36

talking about it a couple years

41:38

ago. So update from a couple

41:41

years ago. Is it still there?

41:43

Are we still worried about the

41:45

Hubble tension? What is it? Hubble

41:47

tension is a big problem. So

41:50

discussed, we have these models, we

41:52

fit for a bunch of parameters.

41:54

to try to explain the measurements

41:56

in the cosmic microwave background in

41:59

galaxy surveys. And one of the

42:01

parameters is the Hubble constant, which

42:03

is the rate. at which galaxies

42:05

are moving apart from each other.

42:08

So essentially measures the speeds in

42:10

which galaxies are moving apart from

42:12

each other. And so, you know,

42:14

the galaxy distribution, the cosmic microwave

42:17

background, we don't actually see the

42:19

universe expanding, but this expansion rate

42:21

is a parameter in the models

42:23

that we describe the distributions of

42:26

the galaxy in the cosmic microwave

42:28

background. But alternatively, you can try

42:30

to measure the album constant directly.

42:32

just like Hubble did 100 years

42:35

ago. So you can look at,

42:37

you know, some galaxies that are

42:39

not too far away, and you

42:41

see those galaxies moving away from

42:44

each other, moving away from us,

42:46

from our galaxy, and if you

42:48

can also figure out the distance,

42:50

then that gives you the Hubble

42:53

constant. So the Hubble constant is

42:55

the ratio of the velocity which

42:57

galaxies are moving away from us

42:59

to their distance. In principle, it's

43:02

straightforward. So measuring the velocity of

43:04

which galaxies are moving away from

43:06

each other from us is actually

43:08

fairly easy. And the reason is

43:11

the galaxies emit light, and some

43:13

of that light is either absorbed

43:15

or emitted by various atomic transitions,

43:17

and so there are spectral lines.

43:20

There are lines in the spectrum,

43:22

the frequency spectrum of the light

43:24

that we see. And if the

43:26

galaxy that's emitting this light is

43:29

moving away from us, then those

43:31

lines are... Doppler shifted to different

43:33

treaties or longer wavelength. So the

43:35

same effect is when an ambulance

43:38

is moving away from you, it

43:40

sounds lower pitch than it does

43:42

when it's towards you. So we

43:44

measure these Doppler shifts, we can

43:47

figure out the velocities very very

43:49

well. The distances are surprisingly difficult.

43:51

And the reason is that when

43:53

we look at a galaxy on

43:56

the sky, it has some angular

43:58

size. And if we knew what

44:00

the physical size was, then we

44:02

could infer the distance. Or if

44:05

we see a galaxy, it has

44:07

some... that we can measure very,

44:09

very precisely. And if we knew

44:11

exactly how luminous the galaxy was,

44:14

then we could figure out exactly

44:16

how far away it was. If

44:18

you give me, if I give

44:20

you a standard flashlight and you

44:23

shine it at me, I can

44:25

figure out how far away you

44:27

are because I know how bright

44:29

the flashlight is and I can

44:32

measure how bright it is. But

44:34

galaxies don't all have the same.

44:36

luminosity and so we can't infer

44:38

the distance by looking at the

44:41

luminosity. It turns out though that

44:43

there are things called supernovae and

44:45

in fact a very specific type

44:47

of supernova type 1A. A type

44:50

1A supernova is a white dwarf,

44:52

an exploding white dwarf. So what

44:54

happens is when a star uses

44:56

up all of its nuclear fuel

44:59

it evolves to a state where

45:01

it's a gravitationally bound star in

45:03

which there's no... nuclear fuel being

45:05

burned. And the star is held

45:08

up from gravitational collapse by quantum

45:10

pressure, quantum electron degeneracy pressure. But

45:12

there's a limit as to how

45:14

massive such a star could be

45:17

before the gravitational forces overcome this

45:19

electron degeneracy pressure. And so... you

45:21

know, if I have a white

45:23

dwarf that's in a binary with

45:26

some other star and it's a

45:28

creating matter from the other of

45:30

the star, as soon as that

45:32

white dwarf exceeds this limit, which

45:35

is called the Shandrasaker limit, it

45:37

explodes. And since that happens at

45:39

a very specific type of mass,

45:41

we believe that all of these

45:44

supernovae are exactly the same. So

45:46

there's a good theoretical reason to

45:48

believe all type 1A supernovae have

45:50

the same. luminosity, and it's also

45:53

been measured empirically. You can look

45:55

at a bunch of supernovae, you

45:57

know, the same galaxy, and they

45:59

do have the same brightness. And

46:02

so these supernovae are what we

46:04

call standard candles. They're objects that

46:06

have... a very well-determined luminosity, and

46:08

so if we observe how bright

46:11

they are, we can actually figure

46:13

out the distance to the supernova,

46:15

and therefore the galaxy that hosts

46:17

it. So there has been a

46:20

project that's been going on for

46:22

15-ish years called the Shoe's Collaboration,

46:24

S-H-0-E-S, and there's also been another

46:26

collaboration called the Caltech Carnegie Chicago

46:29

Hubble Project, CC-C-E-C-C-C-I-O-A-L-O-O-L-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-L-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-O-D They've been measuring

46:31

the Hubble constant in this way.

46:33

They've been looking at supernovae, distant

46:35

galaxies, and measuring the brightnesses of

46:38

the supernovae, and the velocities in

46:40

which the galaxies are moving. And

46:42

when they do these measurements, especially

46:44

as shoes collaboration, they find that

46:47

the expansion rate is about 10%

46:49

larger than the expansion rate you

46:51

infer from the cosmic microwave background

46:53

and galaxy search, and our models

46:56

to account for them. We call

46:58

it the Hubble tension because when

47:00

this was first noticed it was

47:02

a discrepancy but it wasn't clear

47:05

whether it was statistic significant or

47:07

not. It also wasn't clear whether

47:09

it was maybe some misunderstanding of

47:11

how supernovae work or how the

47:14

observations work or perhaps some problem

47:16

with the calibration of the distances

47:18

and brightnesses. But things have evolved,

47:20

you know, with time the measurements

47:23

have become better. There have been

47:25

more of them. and the error

47:27

bars have shrunk and many of

47:29

the systematic effects that people were

47:32

concerned about 10 years ago have

47:34

been shown to be of no

47:36

concern or not, you know, not

47:38

the source of the discrepancy. And

47:41

if anything, things got, you know,

47:43

the Hubble tension became much more

47:45

serious just a few years ago

47:47

with the launch of JWST. So

47:50

until I would say three years

47:52

ago, it was reasonable to be

47:54

skeptical about the brightness of the

47:56

supernov. And the reason is that

47:59

the supernovae brightness is... calibrated to

48:01

things called Cepheid variables, which are

48:03

stars, variable stars. They're stars that

48:05

become brighter and dimmer over timescales

48:08

of, you know, weeks to months.

48:10

And Cepheid variables are also standard

48:12

canned. And so we have Cepheid

48:14

variables in the nearby in the

48:17

Milky Way, and we can measure

48:19

their distances very well. And then

48:21

there's some Cepheid variables in nearby

48:24

galaxy that also hosts supernov. There

48:26

are about 40 such... galaxies that

48:28

host Cepheid variables that we've observed

48:30

very well, and supernovae. So that's,

48:33

so the supernova distances are calibrated

48:35

to the, to the Cepheid variables,

48:37

and the Cepheid variables then are

48:39

what we're, you know, calibrating. So

48:42

when you look at a Cepheid

48:44

variable in one of these nearby

48:46

galaxies, if you look at it

48:48

with the Hubble Space telescope, which

48:51

was the best instrument that we

48:53

had for doing these measure. The

48:55

angular resolution of the Hubble Space

48:57

Telescope is great, but not perfect.

49:00

And in many cases, when you

49:02

were looking at one of these

49:04

Cepheid variables, there would be light

49:06

from nearby stars that would sort

49:09

of spill over onto the Cepheid

49:11

variable limit. So this was something

49:13

that you might be reasonably concerned

49:15

about. Can we actually separate the

49:18

light from the Cepheid variable from

49:20

the light from nearby stars? well

49:22

enough to actually tell how bright

49:24

that Cepheid variable. Yeah. But now

49:27

we have the James Webb space

49:29

telescope that launched what three years

49:31

ago now and the James Webb

49:33

space telescope has much better angular

49:36

resolution than the Hubble space telescope

49:38

does. And so you can go

49:40

back and look at some of

49:42

the Cepheid variables and do the

49:45

measurement with this better telescope. And

49:47

when you do that there's no

49:49

issue of crowding. The Cepheid variables

49:51

are very very well separated from

49:54

the nearby stars. and you can

49:56

make the measurement. And for the

49:58

16 such supernovae, or sorry, 16

50:00

such sefugate hosts for which they've

50:03

done this measurement out of the

50:05

sample. the 45-ish HST, set hosts,

50:07

the measurements are spot on from

50:09

what was inferred from HST. So

50:12

this, you know, crowding, you know,

50:14

issue is no longer a concern.

50:16

And so the Hubble tension is

50:18

more serious now than it was

50:21

three years ago because of this.

50:23

Okay, so there's there. Unless there's

50:25

some huge mistake that we are

50:27

really missing in the experimenters have

50:30

obviously been very good, there's a

50:32

mismatch between the sort of direct

50:34

local measurements of the expansion rate

50:36

and the inferred expansion rate we

50:39

need to fit the data of

50:41

like the CMB and the wider

50:43

models. So at a very broad

50:45

strokes without getting into a model

50:48

building or anything like that, but

50:50

how would we solve this? Could

50:52

we just... What are we looking

50:54

for? Were you looking for something

50:57

that makes the universe slow down

50:59

at later times? Or speeds it

51:01

up at later times? Or what

51:03

is the target here? So the

51:06

target here is the barion acoustic

51:08

oscillation. in the cosmic microwave background.

51:10

So we talked about how you

51:12

see this peak in the correlation

51:15

function for galaxies at 100 mega

51:17

parsecs, but you also see this

51:19

peak in the cosmic microwave background.

51:21

But you don't measure the physical

51:24

size of the peak, you measure

51:26

the angular size. Same thing with

51:28

a pond. If you're viewing a

51:30

pond from some distance and you

51:33

see the ripples going out, you

51:35

see the waves going out from

51:37

where you're the pebble, you can't

51:39

tell how... big those waves are

51:42

unless you know how far away

51:44

you are viewing and viewing them

51:46

from. So we measure the angular

51:48

size of this found horizon with

51:51

varying acoustic oscillation very very precisely

51:53

one part in the 10,000 with

51:55

the cosmic microwave background measurements and

51:57

the angular size is the physical

52:00

size divided by the distance to

52:02

the cosmic microwave background. That angular

52:04

size is determined from cosmological models

52:06

that have as the parameters the

52:09

Hubble constant the dark matter density

52:11

and the barion density and the

52:13

dark energy density and the dark

52:15

energy density and the dark energy

52:18

density and how it evolves with

52:20

whether it evolves with time. So

52:22

the two solutions that people have

52:24

can be sort of classified into

52:27

late time solutions and early time

52:29

solutions. So the late time solutions

52:31

we sort of changed that angle.

52:33

the model predictions for that angle

52:36

by changing the distance to the

52:38

surface of last scatter and that

52:40

would happen if we some if

52:42

the expansion rates in the recent

52:45

past was somehow different than in

52:47

the standard cosmological. Those tend to

52:49

not work because we can also

52:51

get a measurement of the Hubble

52:54

constant from the bearing acoustic oscillation

52:56

of the galaxy distribution and that

52:58

is sort of agrees with what

53:00

we get from the cosmic microwave

53:03

background. So these late time solutions,

53:05

people thought about early on, but

53:07

they don't really work. So the

53:09

other possibility are early time solutions

53:12

where we somehow change the physical

53:14

size of the sound horizon in

53:16

the early universe and ways to

53:18

do that. And in fact, we

53:21

have to decrease the physical size

53:23

of the horizon at the universe

53:25

to account for the Hubble tension.

53:27

And so one idea that... people

53:30

have spent a lot of time

53:32

thinking about these from about 2018

53:34

until now. And I'd say initially,

53:36

it's sort of like a cosmological

53:39

constant, but has a much larger

53:41

magnitude and it's around only in

53:43

the early universe for the first

53:45

half million years of the universe

53:48

and then somehow decays away. And

53:50

people spent a lot of time

53:52

thinking about these from about 2018

53:54

until now. And I'd say, you

53:57

know, initially. We were thought of,

53:59

you know, as a promising idea.

54:01

You were one of these people,

54:03

by the way. I was one

54:06

of these people. And in 2021,

54:08

there was this very interesting result

54:10

from one of the cosmic microwave

54:12

background collaborations, the Act, Atacama Cosmology

54:15

Telescope Collaboration. They published a paper

54:17

late 2021 where they said that

54:19

the new measurements were actually more

54:21

consistent with the early dark energy

54:24

model than the standard lab, the

54:26

CDM. And that was very exciting.

54:28

I'm all excited. And that got

54:30

people looking more revved up about

54:33

early dark energy models. There was

54:35

more a bigger, you know, another

54:37

round of model building, but then

54:39

also more scrutiny from experiments. And

54:42

I think, you know, that's four

54:44

years ago, three and a half

54:46

years ago. And over the past

54:48

three and a half years, we've

54:51

had more measurements from the cos

54:53

microwave background from galaxy surveys, more

54:55

data, more scrutiny of the data.

54:57

And I think the pendulum is

55:00

sweetening back to Lambda CDM. away

55:02

from early dark energy. So it

55:04

sounds like there's two options, late-time

55:06

solutions and early-time solutions, and neither

55:09

one of them work. That is

55:11

a, I would say, fair summer

55:13

of the situation. So it's very

55:15

different, just again, for the sort

55:18

of non-expertz here. In 1998, when

55:20

people claimed, when the two teams

55:22

claimed that the universe is accelerating

55:24

and that was an anomaly, etc.

55:27

There was instantly a theory that

55:29

explained it, and everyone could say,

55:31

oh, okay, so we found this

55:34

thing. Here we have an anomaly

55:36

and it's my impression is it's

55:38

still not clear what could explain

55:40

it. That is correct. I mean

55:43

early dark energy was a really

55:45

a very plausible idea until just

55:47

a few years ago and I

55:49

would not say that it's ruled

55:52

out because early dark energy is

55:54

not a model. It's sort of

55:56

a class of models or an

55:58

idea that can go into models.

56:01

But what's happened is with new

56:03

data. is increasingly more consistent with

56:05

Lamb to CDM. The wiggle room

56:07

for constructing really dark energy models

56:10

has decreased and it becomes harder

56:12

and harder to find some model

56:14

that actually works. So, um... But

56:16

your summary, yes, is probably, so

56:19

the first approximation, correct, the late

56:21

time, the simplest late time models

56:23

don't work, the simplest early time

56:25

models don't work. But it's even,

56:28

it's sort of a, yeah, I

56:30

think if you had Adam on

56:32

your podcast, I mean, what Adam

56:34

would say if he were here,

56:37

is that the evidence for the

56:39

late 90s, but. people are much

56:41

more reluctant to accept this because

56:43

we don't have a model to

56:46

explain it. So if interestingly enough

56:48

it was if it was the

56:50

other way around, so if the

56:52

cosmic microwave background was giving us

56:55

a Hubble constant that was 10%

56:57

bigger than that from supernovae, then

56:59

we would just change our dark

57:01

energy model. Yeah. So we'd say

57:04

it was time evolving dark energy.

57:06

Given that the Hubble constant from

57:08

supernovae is larger than that inferred

57:10

from C&B. The same solution could

57:13

also be tried, but it would

57:15

require a dark energy density that

57:17

increases with time rather than decreases.

57:19

And decreasing with time is okay,

57:22

because the energy can go somewhere

57:24

else. But in order to have

57:26

a dark energy density that increases

57:28

with time is sort of equivalent

57:31

to... having energy just appear out

57:33

of the vacuum, which is not

57:35

something that we like. I think

57:37

about, I think you probably know

57:40

this better than I, and in

57:42

the general relativity community there's, it

57:44

violates the strong energy. It violates

57:46

the weak energy. There we know,

57:49

see, yeah, I think I learned

57:51

this from you. But I did

57:53

write a shape, and I really

57:55

didn't learn it from you. I'm

57:58

forgetting myself, but it violates, but

58:00

your point is right. It's just

58:02

sort of so magical and scary

58:04

to have energy appear out of

58:07

nothing. Yeah, I think that's the

58:09

weak energy condition is general of

58:11

relative for creating energy out of

58:13

the vacuum, which makes us physicists

58:16

uncomfortable. But, you know, the cosmological

58:18

cons and also made us uncomfortable.

58:20

Yeah, I mean, to be fair,

58:22

it's more than uncomfortable when you

58:25

try to construct a model which

58:27

it happens, other things tend to

58:29

go disastrous. I mean, our discomfort

58:31

is not purely emotional and vibes-based.

58:34

Yeah. Okay, I guess I remember

58:36

now, and I know that we're

58:38

running long, so let me know

58:40

if I'm abusing your kindness here,

58:43

but there's another tension, even before

58:45

we get to the variable dark

58:47

energy stuff. There's the S8 tension

58:49

that cosmologists worry about and is

58:52

not sunk into the popular imagination.

58:54

Should we worry about that? Yeah,

58:56

the S8 tension is a little

58:58

more subtle, and I sort of

59:01

have less confidence in it. And

59:03

whether it's a tension or not

59:05

seems to bounce around a lot

59:07

more depending on who you ask

59:10

and which data set. And one

59:12

of the conclusions from the new

59:14

results that we've seen from the

59:16

DESE collaboration and the Dark Energy

59:19

Survey collaboration just last week is

59:21

that with new data the S8

59:23

tension is going away. So the

59:25

S8 tension is. a discrepancy between

59:28

the amplitude of fluctuations in galaxy

59:30

surveys on small distance scales compared

59:32

with that expected from the cosmic

59:34

microwave, the models that best fit

59:37

the cosmic microwave background. And it's

59:39

a strange tension because it sort

59:41

of depends on which data set

59:43

you look at and in some

59:46

cases how you analyze the data

59:48

set. And there's some measurements that

59:50

seem to indicate that there's a

59:52

discrepancy, but then other measurements that

59:55

indicate that it's not a discrepancy,

59:57

and it also involves measurements or

59:59

observations of the galaxy distribution on

1:00:01

smaller scales, where the theory becomes

1:00:04

more complicated and we have less

1:00:06

confidence. So a lot of people

1:00:08

in cosmology... worry about the essay

1:00:10

tension. There's probably less consensus about

1:00:13

whether it's there or not and

1:00:15

I think it might be going

1:00:17

on. Well that's another reason why

1:00:19

the Hubble tension hasn't quite been

1:00:22

as completely accepted as the accelerating

1:00:24

universe because when you find attention

1:00:26

like that, maybe you find other

1:00:28

tensions or other signals that kind

1:00:31

of go in the same direction.

1:00:33

But he, you know, rather than

1:00:35

building up, we're having other things

1:00:37

sort of come and go and

1:00:40

waffle around and nothing quite definitive

1:00:42

yet. Yeah, it's a, it's a

1:00:44

perfectly reasonable view. And I think

1:00:46

10 years ago, most people would

1:00:49

think that, you know, as we

1:00:51

analyze more C&B data, as we

1:00:53

understand the analyses better, and as

1:00:55

we understand the supernova, and the

1:00:58

analyses better, we'll find, you know,

1:01:00

small errors in one or both

1:01:02

that sort of have accrued, you

1:01:04

know, together, give you some consistency.

1:01:07

But that has not happened. Not

1:01:09

yet happened, yeah. Okay, you already

1:01:11

referred to... two new results which

1:01:13

have sadly very similar names. DS

1:01:16

and DESI and they're both attempts

1:01:18

to measure dark energy and they're

1:01:20

both hinting that it is not

1:01:22

to cause molecule constant. So if

1:01:25

true, I'm very, I have sort

1:01:27

of public statements that I'm skeptical

1:01:29

that something like that would come

1:01:31

to pass, but it would be

1:01:34

a big deal if it were

1:01:36

true. I agree. I am also

1:01:38

skeptical. I think sometimes when

1:01:41

I'm skeptical I have to try

1:01:43

to like dial it back. Yeah.

1:01:45

Because I think often what you

1:01:47

find when you study history of

1:01:49

science is that discoveries are made

1:01:51

not just by major fundamental discoveries

1:01:53

are made. not just by people

1:01:56

who've made really good measurements and

1:01:58

are really good at the analysis,

1:02:00

but there are people who have

1:02:02

an open mind and will be

1:02:04

accepting of the possibility that this

1:02:06

might be big. We're not as

1:02:08

young as we used to be.

1:02:11

I mean, if you're, you know,

1:02:13

I think most of us have

1:02:15

this attitude that when there's something

1:02:17

strange in the data, there must

1:02:19

have been something that goes wrong,

1:02:21

went wrong, or we missed it,

1:02:24

you know, the Hubble-people have that

1:02:26

attitude, you know. they're obviously missing

1:02:28

something with supernovae, complicated systems, they

1:02:30

haven't modeled it correctly, the model,

1:02:32

you know, the standard cosmetical model

1:02:34

is fine. And if you always

1:02:36

have that attitude, you'll never discover

1:02:39

anything. Yeah. So what is the

1:02:41

new result? So the new result

1:02:43

is coming from DESE, and then

1:02:45

there's sort of consistent information coming

1:02:47

from DESE and from various supernovae

1:02:49

measurements. show that the dark energy

1:02:51

density is evolving with time. And

1:02:54

in particular, they show that it

1:02:56

is, or has in the recent

1:02:58

past been increasing with time. So

1:03:00

it's a complicated result. It's not

1:03:02

hugely statistically significant than it was

1:03:04

a year ago. But it suggests

1:03:06

that the dark energy density was

1:03:09

smaller early times, became larger. with

1:03:11

time and then fairly recently started

1:03:13

to decrease in energy again. So

1:03:15

it's a very unusual result. So

1:03:17

I would say it's unusual in

1:03:19

several ways. The first is that,

1:03:22

I mean, if the dark energy

1:03:24

density was evolving in time, you

1:03:26

know, that is instant Nobel Prize.

1:03:28

And we've been looking for this.

1:03:30

So, you know, we shouldn't say,

1:03:32

you know, it can't be right.

1:03:34

you know, just discussed earlier the

1:03:37

preferred fit suggests that the dark

1:03:39

energy density was increasing with time.

1:03:41

which I just learned violates the

1:03:43

weak energy petition, which I already

1:03:45

knew is creating energy out of

1:03:47

a vacuum, which is, I'm supposed

1:03:49

to keep an open mind too,

1:03:52

but it's really very, very, very

1:03:54

strange, the point of view, theoretical.

1:03:56

One has priors, that's okay, and

1:03:58

your priors are never zero, but

1:04:00

they're bigger on some possibilities than

1:04:02

others. Yeah. So, I mean, another,

1:04:04

this is, I mean, another way

1:04:07

of saying it's like sort of

1:04:09

higher order in. discovery space, you

1:04:11

know. Learning that the dark energy

1:04:13

density evolved in time is like

1:04:15

spectacular enough. But then learning that

1:04:17

increases time, that's like even beyond

1:04:20

that. So I think the bar

1:04:22

for that is even higher than

1:04:24

it would be just for dark

1:04:26

evolution. What are these experiments? What

1:04:28

is he measuring? So the principal

1:04:30

experiment for this is the DESE

1:04:32

collaboration, which stands for dark energy

1:04:35

spectroscopic instrument, I think. It does.

1:04:37

That's right. I looked at, oh,

1:04:39

so this is a really spectacular

1:04:41

project where they measure the red

1:04:43

shifts and therefore the distances to

1:04:45

millions of galaxies over a huge

1:04:47

volume of the universe. And with

1:04:50

these measurements, they can determine the,

1:04:52

they can measure the berry and

1:04:54

acoustic oscillation feature at a variety

1:04:56

of different redshift or distance spin.

1:04:58

So they can measure the angular

1:05:00

size. of this bump in the

1:05:02

clustering, the galaxy clustering, they can

1:05:05

measure the angular size at a

1:05:07

variety of different distances. And in

1:05:09

that way, they can figure out

1:05:11

the expansion rate as a function

1:05:13

of red shift, or as a

1:05:15

function of time. And so they

1:05:18

can actually see the expansion rate

1:05:20

changing with time in this. So

1:05:22

I mean, some of the issues

1:05:24

are that... They're splitting all of

1:05:26

their galaxy survey into a bunch

1:05:28

of different distance bins, and so

1:05:30

they have bazillion... of galaxies, but

1:05:33

you know, they have six or

1:05:35

seven different distances and so on

1:05:37

each bin, it's a bazillion divided

1:05:39

by six or seven. And then,

1:05:41

you know, the other things that

1:05:43

you might be concerned about is

1:05:45

that, you know, the universe is

1:05:48

actually evolving with time and maybe

1:05:50

there's something about the properties of

1:05:52

the galaxies that they're looking at

1:05:54

that are evolving with time. And

1:05:56

you can read the papers. They've

1:05:58

got hundreds of pages. They've spent

1:06:00

a huge amount of effort. checking

1:06:03

for all these obvious things that

1:06:05

you would check for and none

1:06:07

of these obvious things that would

1:06:09

check for is shown up But

1:06:11

you know with a result like

1:06:13

this That's so unusual you really

1:06:16

require you know a higher level

1:06:18

degree of scrutiny. I mean the

1:06:20

way I look at it I

1:06:22

mean the other thing I should

1:06:24

say is that there are also

1:06:26

supernova measurements that are sort of

1:06:28

like those like the students yeah

1:06:31

but they look at supernovae out

1:06:33

to larger distances. So they're interested

1:06:35

not so much in the expansion

1:06:37

rate today, but how it evolves

1:06:39

with time. So they're doing sort

1:06:41

of complementary measurements. They're sort of

1:06:43

doing the same thing that DESE

1:06:46

is trying to do with the

1:06:48

Bryan acoustic oscillation, but in a

1:06:50

slightly different way. And then there's

1:06:52

the dark energy survey, which doesn't

1:06:54

have distances quite as well, but

1:06:56

they have tons and tons of

1:06:58

galaxies. their measurements are sort of

1:07:01

consistent as well. Consistent with the

1:07:03

DESE results of the time-dependent dark.

1:07:05

Yeah, they can't really determine the

1:07:07

time evolution of the dark energy

1:07:09

quite as well, but there are

1:07:11

other places where their observations overlap

1:07:14

with DESE's observation. And in places

1:07:16

where they overlap, there's consistent. So

1:07:18

I think the way I look

1:07:20

at it is that DESE. is

1:07:22

shown that these galaxy surveys can

1:07:24

be extremely powerful, they can work.

1:07:26

And the other thing that's important

1:07:29

to notice that DESE is not

1:07:31

the last such project. We've got

1:07:33

the Rubin Observatory. that's going to

1:07:35

start taking data any day now,

1:07:37

and they're going to do analogous

1:07:39

things over a few volumes. Ground-based

1:07:41

telescope. Ground-based telescope. There's then the

1:07:44

European Space Agency last year launched

1:07:46

Euclid, which is a space-based, we're

1:07:48

going to do a space-based galaxy

1:07:50

survey. And there's then NASA's Roman

1:07:52

space telescope, which will also be

1:07:54

launching soon, and that's going to

1:07:57

also do a huge galaxy survey

1:07:59

from space. And all these projects

1:08:01

have some overlap, but they also

1:08:03

have complementarities. They check different things.

1:08:05

They will be affected by different

1:08:07

types of systematic artifacts. They have

1:08:09

different ways of observing the same

1:08:12

galaxy populations, and they also have

1:08:14

access to slightly different galaxy populations.

1:08:16

And also two weeks ago NASA

1:08:18

launched a project called Sferex, which

1:08:20

is going to have some galaxy

1:08:22

mapping capabilities. And so the way

1:08:24

that I look at it is

1:08:27

that, you know, these projects can

1:08:29

work. They do work. The level

1:08:31

of precision that we're getting from

1:08:33

them is absolutely stunning and was

1:08:35

unimaginable, just, you know, even 10

1:08:37

years ago. And, and the other

1:08:39

things, you know, the DESE results

1:08:42

are new and... We've always found,

1:08:44

you know, with new telescopes, you

1:08:46

know, projects in cosmology and astronomy,

1:08:48

when you build a new telescope

1:08:50

to make new observations, you're learning

1:08:52

about the universe and the telescope

1:08:55

at the same time. And so

1:08:57

I think it's going to be

1:08:59

really interesting, important and interesting for

1:09:01

us to, you know, really look

1:09:03

at the telescope and the detectors

1:09:05

and the analysis pipelines. simultaneously with

1:09:07

our scrutiny of the cosmological implications.

1:09:10

And I think that in the

1:09:12

process, we'll understand better what's going

1:09:14

on. And even if it's not

1:09:16

time evolving dark energy, it's definitely

1:09:18

going to feed into our ability

1:09:20

to do these measurements even better

1:09:22

in the future. sure some people

1:09:25

are going to want to know

1:09:27

why we need to build so

1:09:29

many different telescopes. Why can't JWST

1:09:31

do this? But these are experiments

1:09:33

designed for different purposes. Yeah, so

1:09:35

JWST is an absolutely phenomenal instrument.

1:09:37

I actually got to see it

1:09:40

in the High Bay at Lockheed

1:09:42

Martin in December of 2019 and

1:09:44

I was looking at and watching

1:09:46

the videos of how it's going

1:09:48

to like unfold and unpack and

1:09:50

I was thinking there's no way.

1:09:53

It's going to work. Like, oh

1:09:55

my God, it was like crazy.

1:09:57

I mean, as if you can't

1:09:59

fix it, you can't go up

1:10:01

and repair it. I mean, the

1:10:03

fact that it worked and actually

1:10:05

worked better than they anticipate in

1:10:08

many ways, it's actually spectacular. The

1:10:10

images, the things we're finding with

1:10:12

it, absolutely amazing. But the thing

1:10:14

is JWST is a narrow field

1:10:16

of view. It's really good for

1:10:18

looking at a very small number

1:10:20

of objects, very large distances or

1:10:23

very faint objects. But if we're

1:10:25

trying to do cosmology, we'll want

1:10:27

to map the decision of galaxies

1:10:29

over as large a volume as

1:10:31

we can, so over as much

1:10:33

of the skies. So it's a

1:10:35

different type of telescope. And one

1:10:38

of the things about the... does

1:10:40

he result is it brings home

1:10:42

at least my very very casual

1:10:44

looking at the papers brings home

1:10:46

the fact that it really does

1:10:48

depend on your model that you

1:10:51

think you're testing when you come

1:10:53

across and say here's our result

1:10:55

right because if you just fit

1:10:57

to there's a constant dark energy

1:10:59

etc. You get one result if

1:11:01

you say well I'm going to

1:11:03

let it vary linearly with time

1:11:06

you get a different result if

1:11:08

I'm going to let many things

1:11:10

happen you get a different result

1:11:12

is it possible that There's different

1:11:14

levels of confidence in the dark

1:11:16

energy used to be increasing result

1:11:18

and the dark energy is somehow

1:11:21

changing result or do they go

1:11:23

hand in hand? I'm still trying

1:11:25

to understand that. So yes, what

1:11:27

we do is we construct mob...

1:11:29

and then we fit for the

1:11:31

parameters in those models. And one

1:11:33

of the things I'm trying to

1:11:36

understand is that if you take

1:11:38

the DESE results and you model

1:11:40

them with the standard Lambda CDM

1:11:42

model, my understanding is that actually

1:11:44

gives you a pretty good fit.

1:11:46

And I don't know whether it's,

1:11:49

I mean, if there was something

1:11:51

that was desperately wrong with Lambda

1:11:53

CDM, then when you try to

1:11:55

fit it with Lambda CDM, you

1:11:57

would get a result that was

1:11:59

not the. you would not get

1:12:01

a good result. But my understanding

1:12:04

is that they do get a

1:12:06

good result with land to CD.

1:12:08

But then, if they expand the

1:12:10

model parameters, say, so instead of

1:12:12

land to CDM, they have this

1:12:14

time-evolving dark energy density. So this

1:12:16

is a model that now has

1:12:19

two additional parameters. It has the

1:12:21

time evolution, and then they have

1:12:23

a second parameter, which is the

1:12:25

time evolution of the time evolution.

1:12:27

And when you do that, that

1:12:29

model seems to provide a better

1:12:31

fit. than Lambda CDM, but I

1:12:34

have not yet really understood whether

1:12:36

it implies that Lambda CEDM is

1:12:38

not a good... Right. And my

1:12:40

understanding is also that if they

1:12:42

just try to fit a model

1:12:44

where you have dark energy that

1:12:47

is evolving with time, but that

1:12:49

also gives you a fit that

1:12:51

is consistent with the standard Lambda

1:12:53

CDM. Okay, is there any relationship

1:12:55

between this result and the Hubble

1:12:57

tension? Yeah, that is a great

1:12:59

question. I mean, as I said

1:13:02

earlier, we're always looking for consistency.

1:13:04

There is no obvious way in

1:13:06

which this connects with the Hubble

1:13:08

to. Okay. I mean, I think

1:13:10

it would be much more exciting

1:13:12

if there was. Yeah. But it

1:13:14

turns out that when you change

1:13:17

or expand the model parameter space

1:13:19

this particular way, it does not

1:13:21

change the Hubble constants inferred from

1:13:23

it. the measurement. It does change

1:13:25

the upper limit to the neutrino

1:13:27

mass which is sort of something

1:13:30

that has been I think one

1:13:32

of the most exciting things from

1:13:34

these results that people have not

1:13:36

been paying a whole lot of

1:13:38

attention to. What is that? Where

1:13:40

does that go? So there's a

1:13:42

neutrino have to do with this.

1:13:45

You haven't even mentioned neutrino yet.

1:13:47

Yeah. No one's been mentioning it.

1:13:49

So you know. in the standard

1:13:51

model of elementary particle physics there

1:13:53

are three different types of videos

1:13:55

electron bion and town neutral and

1:13:57

in the standard model when it

1:14:00

was constructed in the early 1970s

1:14:02

the neutrinos were thought to be

1:14:04

massless and in the standard model

1:14:06

they are massless they have don't

1:14:08

weigh anything but then about you

1:14:10

know 20 something years ago it

1:14:12

was discovered the neutrinos actually have

1:14:15

a small non-zero mass. And so

1:14:17

we know now the neutrino masses

1:14:19

are not zero, but we don't

1:14:21

know what they are. They're very,

1:14:23

very small. So we know that

1:14:25

they're bigger than zero, but they're

1:14:28

smaller than some upper limit. Those

1:14:30

upper limits come from a variety

1:14:32

of accelerated experiments in laboratory experiments

1:14:34

and beta decay experiments. But it

1:14:36

turns out that neutrinos, you know,

1:14:38

the standard cosmological model, predicts that

1:14:40

there should be neutrinos running around

1:14:43

the universe, just like there's... light

1:14:45

and barions, and if the neutrinos

1:14:47

have a mass, then they would

1:14:49

actually contribute something to the cosmological

1:14:51

energy density and affects our cosmological

1:14:53

models. And the measurements that we

1:14:55

have known cosmology are so precise

1:14:58

that the fact that neutrino masses

1:15:00

are non-zero actually has to be

1:15:02

taken into account. And in fact,

1:15:04

with our cosmological measurements, we now

1:15:06

have upper limits to neutrino mass,

1:15:08

which are... complementary and in some

1:15:10

ways better than those that we

1:15:13

have from laboratory experiment. And one

1:15:15

of the things that was really

1:15:17

interesting about DESE is that it

1:15:19

improves the sensitivity to a non-zero

1:15:21

neutrino mass. over what we had

1:15:23

before. So to be clear, are

1:15:26

we saying that they have detected?

1:15:28

Like if we didn't know that

1:15:30

neutrinos had mass, would this tell

1:15:32

us that they did? No, but

1:15:34

they have, so what they have

1:15:36

are upper limits that are starting

1:15:38

to distinguish between the two different

1:15:41

neutrino mass scenarios. So there are

1:15:43

three neutrino masses. We've got good

1:15:45

reason to believe that all three

1:15:47

of them are non-zero. We know

1:15:49

that. two of the masses have

1:15:51

some small mass splitting, and another

1:15:53

pair of masses has a larger

1:15:56

mass splitting, but we don't know

1:15:58

whether how those masses are assigned

1:16:00

to the electron, muon, or tau,

1:16:02

and we don't know whether there's

1:16:04

two lighter states in one heavier

1:16:06

state or two heavier states in

1:16:08

one lighter state. And the DESE

1:16:11

results are starting to say that

1:16:13

the inverted hierarchy, the system with

1:16:15

two heavier masses, is rolled out.

1:16:17

Okay. And this is kind of

1:16:19

super useful. Kind of gone under

1:16:21

the radar in terms of popular

1:16:24

press coverage of the DESE results,

1:16:26

but it's a really, really impressive

1:16:28

accomplishment and could be very important

1:16:30

for elementary particle physics. So it's

1:16:32

a tradition late in the podcast.

1:16:34

We always get to let our

1:16:36

hair down and explore wilder ideas.

1:16:39

So you've already said that the

1:16:41

straightforward... fitting the data implies this

1:16:43

dark energy increasing for a while

1:16:45

in density and then decreasing. That's

1:16:47

already a very, very wild idea.

1:16:49

Are there wilder ideas that could

1:16:51

fit the data? So I'll just

1:16:54

tell the audience like back in

1:16:56

the day when we were younger

1:16:58

than we are now, you were

1:17:00

involved in a couple of papers

1:17:02

establishing the idea of the big

1:17:04

rip as a possible future for

1:17:06

the universe, right? With the dark

1:17:09

energy density, just going crazy upward

1:17:11

in the future in the future

1:17:13

in the future in the future

1:17:15

in the future. and friends of

1:17:17

mine and I wrote papers saying

1:17:19

first. So the technical language we

1:17:22

use for increasing energy density is

1:17:24

W less than minus one. W

1:17:26

is a little parameters, right? And

1:17:28

if W is less than minus

1:17:30

one for dark energy, then the

1:17:32

density goes up. So I wrote

1:17:34

a paper saying, can W be

1:17:37

less than minus one? And we

1:17:39

argued probably not. But then we

1:17:41

wrote a follow-up paper saying, could

1:17:43

you be tricked into thinking that

1:17:45

W is less than minus one?

1:17:47

If gravity, we're changing its strength.

1:17:49

over cosmological time. And we said,

1:17:52

you know, maybe, but it doesn't

1:17:54

look very pretty. Are people exploring

1:17:56

ideas like that? Gravity changing its

1:17:58

strength? I don't know. In the

1:18:00

sense that I haven't seen much.

1:18:02

So in the dark energy literature,

1:18:05

there's sort of like the early,

1:18:07

you know, simplest type dark energy

1:18:09

models, which we called I think

1:18:11

you called them quintessents. No, that

1:18:13

was called. You wrote a fight.

1:18:15

All right. You had quintessents in

1:18:17

the rest of the whole. I

1:18:20

did. I did have that. I

1:18:22

did have that. I helped popularize.

1:18:24

So those quintessents, but then there

1:18:26

was sort of a wave of

1:18:28

alternative gravity models for dark energy.

1:18:30

I have not seen many of

1:18:32

these alternative gravity models showing up

1:18:35

in connection with the Desi result.

1:18:37

But I don't know if I

1:18:39

haven't seen them because they're not

1:18:41

there, I just haven't noticed. They

1:18:43

haven't had that long. We'd like

1:18:45

to think that people take more

1:18:47

than a week to write a

1:18:50

good paper. Well, the Desi results

1:18:52

were also around last year. That's

1:18:54

true. I don't know. I mean,

1:18:56

the crazy idea that I like

1:18:58

to think about, please, is oscillating

1:19:00

dark energy. So there's sort of,

1:19:03

I mean, we know that there's

1:19:05

a cosmological constant now. We have

1:19:07

very good reason to believe that

1:19:09

there was a period that we

1:19:11

call inflation in the very early

1:19:13

universe, which was powered by a

1:19:15

non-zero cosmological constant with a very

1:19:18

large magnitude than decayed away. And

1:19:20

early dark energy also. The early

1:19:22

dark energy models also surmise... that

1:19:24

there's a period of cosmological constant

1:19:26

domination, you know, half a million

1:19:28

years after the Big Bang that

1:19:30

then dies away. And so people

1:19:33

have over the years considered the

1:19:35

possibility that, you know, every few,

1:19:37

you know, logarithmic times in the

1:19:39

history of the universe, for some

1:19:41

reason, there's a cosmological constant that

1:19:43

shows up. for a little while

1:19:45

and then disappears again. A cascade

1:19:48

of dark energies at different times.

1:19:50

Yes. So there were papers where

1:19:52

you would just have essentially just

1:19:54

one quintessence model but it had

1:19:56

instead of rolling down a smooth

1:19:58

hill, it rolled down a bumpy

1:20:01

hill. That would do that. And

1:20:03

then there was also this idea

1:20:05

called the string axi verse. which

1:20:07

was quite popular about 15 years

1:20:09

ago. And the basic idea there

1:20:11

is that in string theory, there

1:20:13

are, in addition to the fundamental

1:20:16

fields responsible for electron and muon

1:20:18

and forks and photons, there are

1:20:20

many many many more fundamental fields.

1:20:22

And there could be hundreds of

1:20:24

them that they call axiom field.

1:20:26

And it is conceivable that in

1:20:28

these scenarios, you could have different

1:20:31

axiom fields sort of randomly becoming.

1:20:33

dynamically important at different periods of

1:20:35

the of the universe. So that's

1:20:37

the thing I kind of like

1:20:39

to entertain. And it's kind of

1:20:41

a, it kind of fits in

1:20:43

to some extent with these DESE

1:20:46

results because these scenarios suggest there

1:20:48

should be some type of cosmological

1:20:50

constant domination that then decays away.

1:20:52

I mean, in these scenarios, whatever

1:20:54

we think is a cosmological constant

1:20:56

now, will then decay in the

1:20:59

near future, you know, several billion

1:21:01

years from now and the universe

1:21:03

will then proceed as if there's

1:21:05

no cosmological stint once again. And

1:21:07

so we're, you know, with Desi,

1:21:09

you know, you look at the,

1:21:11

I told you expand, the dark

1:21:14

energy density is increasing, but now

1:21:16

it's decreasing in time. So it

1:21:18

kind of fits in with that

1:21:20

scenario. The only thing that doesn't

1:21:22

fit in is the increasing density,

1:21:24

which we can't fit or has

1:21:26

not yet been to explain. But

1:21:29

the idea, this is an important

1:21:31

one. we should mention. String theorists

1:21:33

never liked the idea of a

1:21:35

positive cosmological constant. That's hard to

1:21:37

fit into string theory, but zero

1:21:39

or negative, they could make their

1:21:41

peace with. And if the dark

1:21:44

energy is evolving and if it's

1:21:46

decreasing right now, that is back

1:21:48

on the table. We could have

1:21:50

a big crunch in the future.

1:21:52

We could have a negative vacuum

1:21:54

energy at the end of the

1:21:57

day. Yep, that is definitely on

1:21:59

the table. I don't think though...

1:22:01

I don't think that those allow

1:22:03

for the increased density. No one,

1:22:05

no sensible person else for that,

1:22:07

which is, so obviously this motivates

1:22:09

people to really get that right.

1:22:12

Yeah. And the final thing then,

1:22:14

I will give you a chance

1:22:16

to wax eloquent on by refringents.

1:22:18

because there's been a couple of,

1:22:20

a couple of, you know, hints

1:22:22

that maybe there is something funny

1:22:24

going on with the polarization of

1:22:27

light from the CMB. That's, that's

1:22:29

the last anomaly that I'll, that'll

1:22:31

lay in front of you. Okay.

1:22:33

So the cosmic birefringence. I learned

1:22:35

about from a 1998 paper by

1:22:37

Sean Carroll and collaborators. Is that

1:22:39

right? Yeah. Maybe. So your listeners

1:22:42

to know that you wrote this

1:22:44

spectacular paper in the late 1990s.

1:22:46

Where you pointed out that there

1:22:48

may be some physical models in

1:22:50

which... light that has a right

1:22:52

circular polarization could travel at a

1:22:55

slightly different velocity than light with

1:22:57

a left circular polarization. And if

1:22:59

so, a light wave that was

1:23:01

linearly polarized would have a linear

1:23:03

polarization that rotated with time as

1:23:05

a propagator. And that is called

1:23:07

cosmic birefringence. And I wrote a

1:23:10

paper a few years after that

1:23:12

or the year after that showed

1:23:14

how you could test the scenario

1:23:16

by looking at the cosmic microwave

1:23:18

background. the cosmic microwave background, we're

1:23:20

looking at light that's been propagating

1:23:22

for 14 billion years, so if

1:23:25

there's any subtle effect... have more

1:23:27

time to crew in the cosmic

1:23:29

wave background than anything else. And

1:23:31

people have been trying to make

1:23:33

these measurements with cosmic microwave background

1:23:35

experiments since then, and there has

1:23:38

been some hints in the data

1:23:40

that the rotation that the linear

1:23:42

polarization actually does get rotated by

1:23:44

0.3 degrees over 14 billion years.

1:23:46

I think it's very exciting, very

1:23:48

interesting, it's a very very difficult

1:23:50

thing to measure from the data

1:23:53

though. And the primary reason is

1:23:55

that it's hard to calibrate the

1:23:57

linear polarization. So they can measure

1:23:59

differences in linear polarization very well.

1:24:01

So if I give you two

1:24:03

rays of light that are side

1:24:05

by side. and ask you what's

1:24:08

the difference in the linear polarization,

1:24:10

you can measure that very well,

1:24:12

but the absolute linear polarization is

1:24:14

harder to get. And that's once

1:24:16

again, because telescopes are complicated things.

1:24:18

Yes, because telescopes are complex. I

1:24:20

mean, it's not, it's a fairly,

1:24:23

it's just easier to measure the

1:24:25

separation, often between two points that

1:24:27

are nearby than it is to

1:24:29

measure the separation of two points

1:24:31

that are really far away. Okay,

1:24:33

that's fair enough. So it's the

1:24:36

same thing with polarizing. But did

1:24:38

you notice that? Act, the Atacama

1:24:40

Cosmology telescope, also has a tiny

1:24:42

little detection by refrigerants. No, I

1:24:44

had not noticed that yet. So

1:24:46

the result that you're talking about

1:24:48

was from Plunk, you know, this

1:24:51

beautiful all-sky thing. And it's like

1:24:53

marginally statistically significant, and like you

1:24:55

say, it's very difficult, so people

1:24:57

didn't get too excited. But the

1:24:59

Atacama Cosmology telescope, which is a

1:25:01

ground-based thing, you know, they had

1:25:03

a recent data release where they

1:25:06

said, everything fits lamb to CDM

1:25:08

perfectly well, but there is like

1:25:10

a two-point something sigma. detection by

1:25:12

refringents. So I hadn't looked at

1:25:14

those because I was studying the

1:25:16

DESE papers really really carefully in

1:25:18

preparation for this podcast. So I

1:25:21

had to change. to dig down

1:25:23

deep in those papers yet. Well,

1:25:25

yeah, I don't know. That's interesting.

1:25:27

It's interesting, yes. I'll have to

1:25:29

take a look. What are your

1:25:31

feelings? What are your, what are

1:25:34

your, this is where we close

1:25:36

up. Your final thoughts, like 20

1:25:38

years from now, what do you

1:25:40

think we'll have landed on? Most

1:25:42

probably. I think 20 years from

1:25:44

now, we all know much. more

1:25:46

with much more certainty whether the

1:25:49

Hubble tension is real. I'm guessing

1:25:51

that 20 years from now there'll

1:25:53

be new ideas from elementary particle

1:25:55

theory and theoretical physics that make

1:25:57

a phantom energy much more palatable.

1:25:59

Phantom energy is the increasing density.

1:26:01

Yes, W. Yes, increasing energy density.

1:26:04

More palatable to us. And I'm

1:26:06

guessing that we will see dark

1:26:08

energy evolution. Okay. All right. Bowl

1:26:10

in here. Because 20 years, I

1:26:12

picked that because we might both

1:26:14

still be around. Yeah. 50 years,

1:26:16

it's easy to make crazy predictions.

1:26:19

Well, not only will we still

1:26:21

be around, but this podcast will

1:26:23

still be around to be able

1:26:25

to turn it on and say,

1:26:27

look. Okay, so less likely. You

1:26:29

might have to have a revivification

1:26:32

if that's the case. All right,

1:26:34

well, that's a lot to think

1:26:36

about. It's kind of good because,

1:26:38

you know, for a while there,

1:26:40

it was possible to believe that

1:26:42

cosmologists had figured it all out,

1:26:44

right? That we had a theory

1:26:47

that fit the data too well,

1:26:49

but now we're in a more

1:26:51

normal science area where there are

1:26:53

anomalies and we got to bang

1:26:55

our head against them, it feels

1:26:57

good. Yep. It's a, yeah, I

1:26:59

think the frustrating thing is I've

1:27:02

been, as you know, I spent

1:27:04

a lot of time working on

1:27:06

early dark energy, and then I

1:27:08

go and give talks, like I

1:27:10

was invited to give talks all

1:27:12

the time, and at the young

1:27:14

people table, what is it, is

1:27:17

it, early dark energy, I'm saying,

1:27:19

well, The the measurements

1:27:21

we've done in the

1:27:23

the next few

1:27:25

years, and know, know,

1:27:27

if it's really

1:27:30

dark energy, we'll

1:27:32

know, we'll know. But now,

1:27:34

we're like, but now

1:27:36

going like, so

1:27:38

what's going on? say.

1:27:40

I have no idea don't

1:27:42

know what to

1:27:45

say. good for have

1:27:47

no idea out

1:27:49

going There's good for

1:27:51

the young people

1:27:53

out there. There's

1:27:55

still a room

1:27:57

for a really

1:28:00

good idea. All right,

1:28:02

Mark, you mean, right.

1:28:04

Get to work. so

1:28:06

right. for being on

1:28:08

thanks so much

1:28:10

for being on

1:28:12

the podcast. Thank you

1:28:15

very much for

1:28:17

inviting me. It's

1:28:19

been an honor

1:28:21

and a joy

1:28:23

to be here.

1:28:26

podcast.