I'm here to shoemaker on the director of
the center and we're really delighted to

have George F. stop you here tonight with
us it's a rare pleasure to have someone of

such free now nd reputation on cosmology
to visit Atlanta and Georgia Tech his

first time in Atlanta for actually first
time in Atlanta Airport How rare is that.

This is really the special thing where
we're glad you got to visit the airport.

He holds the professor of astrophysics
at the University of Cambridge He's also

the director of the Kavli Institute for

cosmology in Cambridge for
those of you who don't know Kavli.

He passed away unfortunately famous for
seeding these wonderful institutes of

physics and other things around the world
so it's a very special place he comes from

his pastor is an Oxford he told me not to
do everything so I'll shorten it is beach

day in Durham University he actually was
in the US for a little while as a post doc

at Berkeley so got a little United States
in there before returning to Cambridge.

And he's had his career mostly at Oxford
and Cambridge ever since then and

I mean Cambridge in the U.K.
so that you don't get confused so

he's won numerous awards I'm only going to
list a few that I can pronounce the one

nine hundred ninety Maxwell medal in prize
of Insta physics the two thousand and

five American Institute of Physics Hyman
prize for Astronomy two thousand and

eleven Gruber cosmology prize
both of those joint and

the twenty thirteen allied Nimitz us
prize that was a little beyond me he's

also a member of the Planck satellite so
please join me with a warm welcome.

Thank

you.

I think this is working well
thank you very much for a very

kind introduction it's
a real pleasure to be here.

So the total of the talk is
the birth of the universe.

I'll tell you.
Bottom line now which is that by the end

of this talk I want to convince
you that we have learned something

about the physical conditions
within ten to the minus thirty five

seconds of the origin of the universe and
that this is really reliable so

we have we really have learned
something about what went on

very very close to
the birth of the universe.

So let me start with
a bit of introduction.

What we're trying to do in theoretical
physics is to make a theory of everything

based on three principles are on gravity.

Signified here by Newton's gravitational
constant by the principle of relativity.

That signified here by
the speed of light giving us

the maximum speed for any physical object.

And quantum mechanics and

his Max Planck and the.

Planck constant that separates the micro
physical quantum world from the classical

world that we're familiar with so
that's what we're trying to do.

And that immediately raises some
very interesting childlike problems.

So this was actually noticed by
Planck assume this Planck introduced

the Planck constant because he
realized that from these three.

Physical constants we can define.

The A characteristic length mass time and
energy and that these are the fundamental

units of physics OK we don't
call these units Planck units so

to theoretical physicist Planck's
constant speed of light and

you can set them all to unity.

Because of the so.

And then if we express these
these characteristic lengths

of fundamental physics mass and
time in the units that we're familiar with

then we get funny numbers.

OK so our so here.

Are is the length expressed in meters so
these units are chosen for

our convenience as human beings OK So you
know a meter is about the length of a long

step in a second it's about the time
interval between heartbeats and

the kilogram is about as much
Pastor as anybody can eat OK And

so then we get funny numbers
the characteristic length scale

the fundamental physics is ten to
the minus thirty five meters which is

you know very very much smaller
than the scale of the universe.

The characteristic time scale the Planck
time is ten to the minus forty four

seconds I universe is thirteen
point eight billion years old.

OK.

You know the total Are you
took out a characteristic

our energy scale is ten
to the nineteen G.V.

which is enormous compared to the present
energy scale of the universe which is

around a million light Transvaal OK so
you get these funny numbers so

somehow from you know if
our tent to make a unified

theory based on these fundamental
principles of physics is to work.

These numbers describing our
universe our natural numbers

have to come out of the theory.

OK so
that that is the theoretical problem.

OK so we can ask.

You know these childlike questions
why is the universe so big.

Are an associate of the related question
which is why is the universe so all.

We see a universe that is
very nearly homogeneous and

isotropic in other words it's
uniform on the large scale.

But it's not completely uniform and

isotropic it has small fluctuations that
have grown to make all of the stuff that

we see in the universe today the planets
and stars and galaxies and ourselves so

we need to understand why is the universe
nearly but not completely uniform.

And then we can go to ask even deeper
questions you know what happened in

the beginning what gave birth to the Big
Bang what was the physics responsible for

that and we could ask questions
which you may think a metaphysical

book by the end of this talk us think you
will I will make the case that this is

actually part of physics of
are there other universes.

OK so how can we tackle these problems so

the this is probably the most conceptually
difficult part of this talk so

this is a space time diagram
in which we're in the middle.

Now because.

It takes a finite time for
light to travel.

A given distance as we look at
distant objects we see them

as they were at an earlier time
in the universe's history so

what's shown here are the paths
of two light rays.

Coming towards us.

So.

So if we track the path
of a single photon.

Then at very early times
the universe was hot and

the hydrogen in the universe
would have been ionized.

Now this happens.

When a when the universe was around
four hundred thousand years old.

So at those early times.

The matter in the universe was
highly ionized and so a photon.

Would be bouncing out from electrons
via Thomson scattering and

he would have very small mean free
path and so he just jiggle around and

then there's the universe expands the.

Temperature of the universe drops protons
and electrons combine to make hydrogen

the universe becomes
transparent to radiation and

the radiation travels
towards us in straight line.

And so if we look in a particular
direction at the sky and look at it.

The the radiation field then you
know if we have an over dense region

then a photon will be covering the red
shifted as it climbs out of the potential

in the evidence region we haven't under
dense region over here then it would be

gravitationally blue shifted so that
means that we can we would expect to see

differences in the temperature
of the background radiation.

And this is just the remnant radiation
from the Big Bang that tells us

the differences in the gravitational
potential in different directions or

on the sky.

So arm.

So we can actually measure these.

Are temperature differences.

Are so the temperature difference
is about equal to the potential

gravitational potential difference
divided by the speed of light squared.

And then if we do a little
bit of basic algebra and

what I've written here is assumed
that the amplitude of the density

fluctuation scales as a power of
the length scale is the fluctuation and

so we would expect to see temperature
fluctuations that scale like Lambda

parameterized the scale
dependence by a parameter and S..

OK so.

If this equation of fairness is one.

Then the fluctuations
the independence of scale and

in the early days of cosmologists.

Because apologists used to assume that the
fluctuations these potential fluctuations

would be independent of scale and
the reason for that was because.

They couldn't conceive of
a physical mechanism that

it would introduce some
characteristic scale.

You know relevant to cosmologists they
just made that basic assumption as as

we'll see that is wrong OK and
it's wrong for a very good reason so

by measuring temperature fluctuations in
the remnant radiation from the Big Bang

we can infer something about
the potential fluctuations.

And in general relativity the potential
fluctuations once they're

generated then they froze so.

So by measuring these fluctuations
you can infer the potential

fluctuations at the time that the
fluctuations themselves were generated and

that's what gives you some leverage
to you know from observations

of what's happening when the universe
was four hundred thousand years old so

inferring what happened
a very much earlier time is.

OK now you could also see on this
picture that there are little ripples.

Fluctuations in the the matter and

the radiation on small scales
oscillate like sound waves.

And they the effective speed of
sound is nearly the speed of light.

And so we also expect to see signatures
of these sound waves on small scales

in the background radiation and
the precise details of those

fluctuations depend on the effective
speed of sound and that.

Depends on the matter counter so again by
looking at the temperature fluctuations

on small scales you can get a handle
of the composition of the universe.

OK so the background radiation
the remnant radiation from

the Big Bang was discovered in one
thousand nine hundred sixty five and.

Immediately following that discovery.

Because knowledge ists started.

A campaign to try and detect these
temperature fluctuations OK and.

They weren't detected until
nine hundred ninety two

when the results from the kopi
Nasa's Coby satellite

were first published and this shows a map.

Of the temperature fluctuations
the typical amplitude of the temperature

fluctuations is a about
a thousandth percent so

this background radiation now has a
temperature of two point seven Kelvin but

with fluctuations.

Of a thousandth of a percent.

So there are some interesting in basic
things that you can infer from this matter

so if you look at this map you can see
that there are large scale correlated

structures here.

And you know in fact the resolution
of Kopi was about seven degrees to

the typical resolution element is about
the size of one of these little blob so

you can see that there
are structures that are much bigger

than the resolution element of Kobe.

OK so.

Is this important.

OK Well you see here's
an interesting thing.

So we see these correlated
fluctuations and

presumably they were generated in
the early universe but this is a.

Of the temperature field as it was
when prostitutes and electrons

combined to make neutral hydrogen so
it's a picture of the universe as it was

when the universe was four hundred
thousand years old so the maximum.

Causally connected patch in this map
is just the speed of light times

four hundred thousand years old
the four hundred thousand years and

that would suggest and.

An angle of about two degrees on
this map and other words these.

Patches these correlations
these correlated structures

were not in causal contact at
the time that we see them.

So whatever physics was responsible for
generating these fluctuations

apparently violates causality.

OK so that's yet another interesting thing
that we have to explain the physics that

generates these apparently
violates causality.

OK so that's a pretty important discovery
you know when people talk about you

know is there evidence of physics beyond
the standard model of particle physics

you just have to look at this map like
this very very exotic physics involved

in explaining this man and so

you know this was recognized
with the two thousand and six.

Nobel prize to the P.I. the two principal
investigators of Kopi George Smith and

John Mayer.

OK So we do have a theory.

And this theory is called inflation and
here's a cartoon picture of.

Of inflation so
the idea is that if a very early times.

The universe was full of stuff
that had negative pressure.

So that equation of state
where the pressure is

approximately equal to minus the density.

Then space itself expands
faster than the speed of light.

And so you can imagine a small
little patch of the universe perhaps

a small as a suppliant playing.

Your head to the minus thirty five
centimeters and then the the universe

is full of this stuff space expands
faster than the speed of light and.

You can produce from this very small
patch a course of a connected region of

the universe that is many times bigger
than the observable universe today

I mean by many times it could be one
hundred orders of magnitude bigger two

hundred orders of magnitude bigger
depends on the details of the model so

that's how you can get a small
you know from fundamental

physics that involves these are natural
numbers that's how you can produce

a very large universe today.

So this the basic ideas behind.

It the inflation in the inflationary
model were developed by Alan Guth Linde

sloven Morgan of our star been
ski pole Steinhart amongst many

other people so
that's the cartoon picture.

And so you know if we imagine.

A spherical ball that we expand
by many many orders of magnitude.

Then the surface will look specially flat.

Because it will look like a flashy and
similarly if we take a region.

Of the universe and inflate it by many
many orders of magnitude we would expect

the geometry to be spatially for.

Our.
So can can this be realized.

And it's quite easy to construct schematic
models for how this might happen so

this is a sketch of the simplest model it
can happen if the universe or early times

is dominated by a scalar field
rather like the Higgs bozo's.

OK so the field Fi This is
the amplitude of the field and

on this axis it's the the potential
energy of the field So

here I've just drawn this relationship
between the potential energy and

the value of the scalar field as
a parabola OK we'll see that we could

actually experimentally constrain
the shape of this potential.

From plaque OK so the dynamics
in this situation is in a very

much like what would happen if you
rolled a ball down a curved surface

if we start up here the ball will
move down to the minimum and

then just oscillate around
the minimum OK so in.

In an expanding you scenario what actually
happens is that the ball will roll down

the universe is expanding and
the axis of frictional force and so

this slide down this potential
is actually very slow.

OK And if it's very slow.

And the ball is moving very slowly
then we get an equation of state

with negative pressure and
the universe expands our exponentially.

Eventually the ball does
actually speed up and then ends

and the ball then oscillates
around its minimum and

if this field if the scalar
field disc up bald to

other particles then you
generate the particles.

The standard model.

And so the oscillations will get damped.

And you create the matter of the Big Bang.

So all of this our inflation period

happens before the matter
in the universe is created.

So that's how we can realize it.

So here are some
predictions of this model.

The universe should be spatially
flaps of very high precision over.

The year because of this
expansion by many many orders of

magnitude the universe should be
homogeneous and isotropic on large scales.

But.

I describe the classical
trajectories quantum mechanically.

The scalar field is fluctuating and
those fluctuations introduce

are fluctuations in the special curvature
of the universe and those fluctuations

then produce the structure that we see in
the universe today so this is a very in.

Beautiful mechanism for generating
the structure that we see in the universe

today the universe on the large scales
is how much we miss an isotropic but

we can generate small fluctuations arm and

those fluctuations you know the galaxies
that we see in the universe today

the groupings of galaxies and so
on were caused by quantum fluctuations

The becomes stretched in scale by
very many orders of magnitude and

become classical fluctuations
in the curvature.

So we would expect to see I won't tell you
technically what an a diabetic spectrum

is but we expect to see a spectrum
of fluctuations and if you remember

I said that in the early days of because
biology because biologists thought that

the fluctuation should be scaling Berent
because they couldn't think of a.

The reason for introducing a scale.

Relevant because Mahler G.
Well in inflation.

There is a scale at some point
inflation has to end and that.

Pulls the spectral index to be
slightly less than unity so

detecting as a trend.

Of the spectrum of fluctuations
the scale dependence of the fluctuations

is a signature of a scale that.

Is related to the end of inflation.

OK so.

So while the scalar field is
moving slowly in this potential

the universe expands exponentially.

H. is the expansion rate
of the universe and

so that is a nearly
constant during inflation.

And.

And is equal to.

Basically that the our whole parameter.

Is the square root of
the energy scale of inflation.

And we also expect quantum fluctuations
in the curvature itself and

that will generate the spectrum of
gravitational waves that again we can

detect through our observations of
the cosmic microwave background and

the amplitude depends on the energy scale
of inflation so if we detected these

gravitational waves we would know the
energy scale of which inflation happened.

OK So when do we think inflation happened

well here is the Atlas
detector from the L.H. C. and

I updated these numbers at the weekend
because as you know the L.H. C..

Had its first successful.

Circulation of a beam.

At its.

When you need to revise
form with the target energy

of fourteen terror Trumbull's
fourteen thousand.

G.V..

So the energy scale of inflation
we know is bounded to be

less than ten to the sixteen G.V..

So that's an experimental bound and

we believe that it happens
close to the center G. scale.

So the physics that we're
talking about is physics that

is operating at energies that
are twelve orders of magnitude higher

than the energy scale of
the Large Hadron Collider.

So I think that you know it's important to

bear in mind that you know
if we are extrapolating.

Our understanding of our physics
in fundamental particles

by twelve orders of magnitude then we
really are into a speculative domain

OK we're very far removed
from experimental tests

which means that any
information that we can get.

Is really valuable information
because it's telling us about physics

a very very high energies.

OK And to put it into.

Another way.

At the end of inflation at the end of
this inflationary period if you took

everything in our
observed the universe and

packed it into a volume so
one of the billions of galaxies that

is seen by the Hubble space telescope and
so on then the entire observable

universe would fit into a volume
smaller than this grapefruit.

OK So at the end of inflation
you could you know theoretically

hold the entire observable
universe in your hand.

So it's a very big extrapolation now
because of that big extrapolation.

We do not have a real theory of inflation.

OK so what we have are.

You know.

Theorists examples of how
inflation my actually work but

we don't have a fundamental
theory of inflation so this is.

Our.

The transparency from.

That was compiled by one of my colleagues
in Cambridge Porsche Ellard and

he just searched in the literature for
papers with the inflation in the title and

then made an alphabetical list of all of
the inflationary models that people come

up with and bear in mind that this is
the first of five transparencies OK So

there are many many models of
inflation that people have proposed.

And that's not surprising because
the physics is very speculative So

if you want an easy career in theoretical
physics you know you can make models of

inflation OK I think I think
that is really unproductive

work OK so what I want to talk about for

the rest of this talk is how
can we actually constrain or

learn about the dynamics of inflation
about what actually really happen.

OK so we've been very fortunate in in
this subject because in each decade or so

this is my history of observational work
on the microwave background radiation.

So we've had three in each one of the.

Issue of the last ten years we
have had a major satellite project

dedicated to measuring.

The the temperature properties of
the cosmic microwave background so

Cobi that I already showed
you which actually discovered

temperature fluctuations in
the microwave background was low.

In one thousand nine hundred nine and
then in two thousand and one the W.

Map satellite was launched and
then in two thousand and

nine the Planck satellite was launched.

OK so here is the picture that
I showed you earlier are the W.

about satellites had a resolution of
about fifteen minutes compared to those

seven degrees of Kopi.

And that they produced this
back in two thousand and

three.

So you know I was with I've been
working on the plank satellite.

And we first proposed the Planck satellite
in one thousand nine hundred two.

OK but at the same year that
the Coby announced the.

The discovery of temperature fluctuations
in the microwave background.

So this was a sketch.

Of of the proposal you
know it was a the short

proposal to the European Space Agency.

And that was one of nearly seventy
missions that was selected to go on for

subsequent investigation and then that
number was cut down to six missions.

And then in one thousand nine hundred
six ease a European Space Agency

are selected one out of the six
missions so they've already spent.

We our team and already spent four years
working on this project I can tell you

that the it's very very tense and
right when you go through one of these

selection processes are but
fortunately Planck was selected and

then it was a long time until
May two thousand and nine

that we actually launched the satellite so
these projects take a long long time.

And you know when I first started working
on Planck I was a much younger and

slimmer man.

So here is an artist's
impression of plank.

As it was when when we flew.

OK so.

So it was launched from French Guiana.

From Korea and I was fortunate
enough to to go to the launch

how many people here have ever been
to the launch of a big rocket.

That's quite it's quite
a reasonable number actually.

So those of you that haven't
if you get the opportunity

then I really recommend OK it's.

Very very different to
actually be here at the launch

you know compared to you
know just watching it.

At the T.V. on the on the T.V. monitor.

So so it was launched on a heavy lift
version of the Ariane five rocket and

it was actually launched together
with the Herschel Space Telescope and

these were views.

From the position that we were out with or
about

eight kilometers away from
the launch satellite.

So it was launched now I'm
really a theory just OK and

it is very easy to make a fool of
yourself for a launch OK because

here is the launch and
then this is what I saw.

Now that doesn't look good.

OK So
you know you've waited all these years for

the launch of this satellite and
then you see this.

But this is this is perfectly normal OK
this is just an optical illusion it's just

you know this is just a quiet enough
altitude that it's starting to follow

the curvature of the earth and so it looks
as though it's falling down so it's the.

You know so I was panicking.

And you know people who were experienced
that Laci said What's the matter with you

you know.

So everything's normal and I think
this is really a great shock because.

So this is our twenty's about twenty
seven minutes after launch this

Africa are in the background here and
this is from a camera on board

the Herschel Space Telescope Herschel is
a three point six meter infrared telescope

which the engineers thought was a good
idea to sit on top of our little.

Satellite and our little
satellite is in this shroud here

and so
at this point Planck was detached from

Herschel then the shroud was blown off and

then Planck started.

Its journey to the out of the ground
point of the Earth Sun system one

point five million
kilometers away from Earth.

Which took a few months.

And it is you know just remarkable
that the the the accuracy of these

launches because.

You know there was a amid course maneuver
scheduled to get plank into the it's

right orbit that I'll to which was
unnecessary because the launch was so

precise so once it got to.

Its this out of the ground in
point it then scanned the sky and

this is what Planck produced.

It produced maps of the great background.

At nine frequencies so
it had two instruments on board

are a low frequency instrument the covered
the frequency range thirty forty four and

seventy gigahertz.

And then it flew meters
long into detectors.

That are come.

Frequency range hundred hundred forty
three to seventeen three five three five

four five and eight five seven gigahertz
are all of these detectors would try

gently cooled and the below meter
detectors would call to point one Kelvin

and so plant was actually a very
complicated satellite with a.

Complex cooling chain.

And so it was really sort of.

You know the engineering behind it was.

Was you know really innovative and
quite risky.

So plank is the you know the cold
the coldest I think my colleagues say

is the coolest satellite this ever flown.

But yes so so these all of these.

Had to text as we're cryogenically called
and we had polarization sensitivity

on all of these channels except
these two highest channels.

OK So what can you see from these maps.

So that these frequencies.

I think you can see straightforwardly
that are above the galactic plane.

There are fluctuations that
are independent the frequency those

of the fluctuations from.

The early universe.

At these high frequencies what
you're seeing is dust emission.

And you know there's no because
cosmological information from here these

channels were included on Planck
to our monitor galactic dust and

when you have this frequency coverage
subtracting out the galaxy from

the cosmological fluctuations is
really very very easy when you can.

You know you can get publication quality
results with a few minutes of computing.

That's not a problem and so here is the.

From the map that was
produced by Planck for

the temperature and I saw please.

And so what do we do with this
map where we try and extract.

Try and quantify how much information
is contained in this map so

I think you can see from this is
a picture of the sky from from Cambridge.

And so
I'll just ask the question you know.

Which of these two pictures contains
more information that one or the.

OK So I think that the answer
to that is self-evident right.

And so that's what we're doing
with with this man OK we try and

see how much information
is contained in the map and

that's a very important question because
every piece every new piece of information

we can get out of this map tells us
about the physics of the early universe.

And so how do we do that where we
can measure various statistics.

So we can measure for example

the probably the joint probability of
finding a temperature difference delta T.

one separated by an angle
theatre from another point

with the temperature difference delta T.
to that is a two point statistic and

we can do the same for a triangle shapes
or for quadrilateral and so on and so

we can make a hierarchy of statistics and
if we free transform them

the correlation functions then
we the two points to distinct.

Tells us about the power spectrum of
the fluctuations the three points distinct

about the by spectrum of the four point
about the tri spectrum and so on so

we can do these manipulations on the map
and try and measure these different

spectra So most of what I have to
tell you about is based on two point.

Statistics.

And so here is the power spectrum from

Planck we published the first
results in twenty thirteen.

So the way to read this is this tells
you the mean square temperature

difference as a function of angular scale.

OK so
our scale is measured here by multiples

which is the inverse of the angular scale.

Expressed in radiance So

a multiple of two corresponds
to ninety degrees of separation.

One degrees about two hundred.

Thirty Arminius here and five Batmen is
close to the resolution image of Planck.

You know is multiple about
two thousand two hundred.

OK so what you see here is that
the are that the amplitude of

the fluctuations in space is independent
of scale large angular scales

then you can see a series of peaks OK And
this scale of this peak

marks the maximum diffs distance that the
sound wave can travel from the beginning

of the universe to when the universe
was four hundred thousand years old and

so these are the signatures of sound waves

in the cosmic plasma and
the independence of scale here

at large angular scales tells us the
fluctuations of nearly scale invariant.

OK And that fits the data.

Essentially perfectly the blue points
here are the measurements are averaged in

bands of multiple that have a with
the Delta L. of twenty five and

they have Arab bars on them and
you know the the.

Level of agreement is so

good that this kind of picture is
an informative and so are shown the visit.

Jools with respect to the best
fit theoretical model.

So our claim in twenty thirteen
was that inflation gives

an essentially perfect
description of the state we.

Are then in twenty fifteen
February we published our

results based on the full Planck mission.

So that there's an improved
signal to noise we also used

more of the sky in this analysis and
you can see that.

All that's happened is.

That the the jiggles
the error bars have shrunk.

The chick was in the points reduced and
the theory fits

the data Senshi perfectly.

OK so if we track our how these
measurements of evolved over time.

Then.

This is what measurements
of the power spectrum from

are a whole bunch of ground based
experiments the spot they look like.

In one thousand nine hundred nine.

And then in two thousand and three now
if you had these black points here

that the first results from the satellite.

And then the situation before Planck flu.

And then if you have seven
hundred million euros to spend.

Then you can get that.

OK.

Now it's it's actually
unusual in physics to get

a theoretical model bear in mind that
the theoretical predictions here were

computed you know twenty years before
these measurements were made OK

to get a theory that fits
experiment this well as this OK So

I think that you know
it really is you know.

We have discovered of you know of

a fundamental truth about
how we our universe.

But in addition we measure polarization.

And.

And so here a now polarization
spectra this Bactrim

is the cross correlation between
the temperature maps and

the polarization maps and I'll tell
you what the stands for later for

the moment is just polarize ocean OK So
these are the measurements.

OK And this red line here is
not a fit to these points.

It's what we predict from
the fit to these data.

OK And it provides Senshi perfect fit and

then this is the polarization
polarization spectrum.

And again the red line is.

It's not a fit to these points it's
predicted from the temperature analysis.

So we have you know very
good agreement with.

You know between the temperature results
and the polarization results and for

those of you that know something
about inflationary models.

This level of agreement you know tells you
that the fluctuations are really very very

purely a diabetic fluctuations OK.

And in addition you know what I was
telling you about light rays travelling

in straight lines towards
this that's not strictly true

because there's matter in the way.

And that matter of course is
the small deflection just through

your gravitational are the growth
of the gravitational effects.

So that intervening matter
distorts the pattern

of the cosmic microwave background and
we can measure this distortion.

So very beautiful analysis because
the way that we measure this distortion

caused by gravitational lensing of
intervening matter is through measuring

the four point statistics OK so
it's a nicer analysis and

this shows the hour
twenty fifteen matter of

the gravitational lensing potential and
most of the features that you see.

On this map a real features and
and what is telling you

is the face of all of the matter in
the universe along the line of sight.

OK on what that has done to

the cosmic microwave background.

And so we can calculate
the spectrum of the lens in a.

Potential field.

And these are.

So the boxes around
twenty fifteen resolves.

The These are orange points
from the twenty thirteen

analysis and their results also from
ground based experiments here and

so our analysis is of you know forty
Sigma detection of lensing and

the the line is what we would expect
from the fit to the temperature.

Data from plaque.

Because it's not a fit to these
points again this is a prediction and

this is important because.

You know we're we're measuring

the temperature fluctuations as they
were when the universe was four hundred

thousand years old that
a redshift of a of a thousand.

The lensing.

Is coming from typically from
matter to redshift of two.

So this.

This model hangs together right
the way to redshift or to.

OK.

So.

You saw that in the background radiation
there were peaks caused by sound waves.

Are those sound waves also generate
imprints a pattern on the distribution

of dark matter and this is been measured
using galaxy redshifts surveys.

So you use galaxies as traces
of the dark matter and

then you measure that they're spectrum.

And these are the best results to date.

So so this is the power spectrum of the
galaxy distribution as a function of wave

number and you can see very clearly and
the syllabary power.

OK we can use the facilities Patton
you know very much like the we

use the cosmic microwave background
to give us a standard ruler and

then we can use it to
test because Mahler G..

So these are still three
patterns are called Barry and

acoustic are salacious and they've been
measured in a number of surveys and

here's how they match the results
from the Planck satellite so

R.S. is the year.

That the sound horizon that you
measure from the location of those

because to features and then D.
is the distance to the galaxy survey so

what I've shown is this ratio measured
from the various galaxies surveys

divided by what you would expect to find
from the best fit model to Planck and

so the Planck best fit model so
if everything was in agreement.

All of the points should lie our
unity here are the Planck plus or

minus one sigma as shown by this band and
then the.

Observations shown here so we get
really good agreement with that data.

Now since I'm running out of time.

Pretty catastrophic Lee.

I will skip our say something about this.

Combining the cosmic microwave
background radiation.

Data with these measurements of
acoustic features in Galaxy surveys

gives us a very precise constraint on
the spatial curvature of the universe.

OK if the universe is spatially
flat if the geometry spatially flat

you should lie on this dotted line.

OK now you cannot do this precisely with
observations of the cosmic microwave

background on their own but
you can if you combine the.

My quick background results with
the galaxy survey results and

that gives us this little allowed
region here OK which tells us that

the universe is spatially flat to
about a precision of half a percent.

OK So that prediction from inflation
is verified to very high precision

OK so I want to skip to this.

Which is a shame because there isn't
jokes in here and I wanted to go to.

Two planks and gravitational waves.

So what I plotted here is
the at a possible amplitude.

Of a of gravitation waves
generated in inflation so

the number are tells you that the
fractional contribution of gravitational

waves to the temperature power
spectrum that we measure.

So we can use the Planck certain set
limits this is the spectral index for

the fluctuations along this axis and

we can use the Planck data
to set limits on this and

so here of an hour just concentrate
on these blue lines you know one.

To sigma or account yes OK so
you can see that from Planck

the two sigma upper limit
is actually point one two.

From our these measurements but this kind
of analysis is very model dependent so

if I you know modify the neutrino
section of the theory then.

The allowed regions shifts to these great
concerts but that's what we got from.

So then OK.

So there isn't another way of testing.

For gravitational waves that is
our less model dependent and

that is to look at the arm.

Or.

At the character of the polarization
pattern so fluctuations that

grow to make our galaxies our producer
polarization pattern that looks

like an electric field that's why we call
it an emotive polarization pattern but if

you have gravitational waves gravitational
wave perturbations have a different

parity to scale of fluctuations and
they generate a curly type pattern

that looks like a magnetic field if you
detect this signature it cannot be sourced

by scalar fluctuations it has to come
from gravity waves so if you make

sense of polarization measurements
you can test gravity waves and

as you know back in March last year
the bicep team produced this image so

this is the bicep map
is beautiful experiment

very high sensitivity sampling
a small patch of sky OK and

they found are a highly
significant detection of this B.

mode power.

OK And this shows their spectrum.

OK And the thing to concentrate on is.

Black points.

OK.

And this dash line is
a theoretical model that has.

Twenty percent of the anisotropy is
coming from gravitational waves.

So tense a scale ratio
of point to this line.

The solid line here is
what you would expect

if there were no gravitational waves and
you do expect to see a little bee mode and

I saw trippy even if there are no
gravitational waves because lens in

whose is gravitational lensing by
intervening matter whose is a small B.

mode signal but
it was this excess that fitted beautifully

with the gravitational wave spectrum that
led the bicep team to claim that they

had this detected gravitational
waves from inflation.

And if that had been true it would have
been a very important result because it

would have stablished the energy
scale of inflation and

it would been the first example of
a physical effect that really does

need quantum gravity to explain it so
a very significant discovery and

I was stunned by this because they
were planing a signal up here.

Which conflicted with what we.

Were claiming now our constraints
a model dependent but

if you if this were really true then you
not only needed gravitational waves but

you needed some other
new physics to reconcile

results from the two experiments.

So this was the and also on the B.B.C.

website generated a huge
amount of publicity.

And then by January.

It had gone OK and it went because and
literally this this.

It took us just a few minutes to
realize there was something really

seriously wrong with the bicep claim.

These are maps polarization
maps from the Planck satellite

three hundred fifty three you go it's.

And this is the bicep region and so

these are color coded by
the amplitude of the polarization.

You know polarization measurements at
these frequencies Now these frequencies

there's nothing cosmological here
it's all coming from polarized

emission from galactic dust.

OK The the bottom panel So this shows the
northern hemisphere southern hemisphere

bicep to region is around the useful
region is four hundred square degrees.

And and so you know it's they were not
looking in the clean region of the sky and

so this raised the to use the alarm
bells that maybe what they were seeing

was entirely due to collecting dust but
this was not good enough it wasn't good

enough from just the plant data to be able
to say for sure that all of the signal

that they were seeing was from galactic
dust and the reason is because Planck

doesn't have a lot of sensitivity
on such small patches of sky.

So what we did was we we joined forces and

we arranged a collaboration if you
want to improve the sensitivity

on galactic dust you have
to directly cross correlate

their maps of one hundred fifty gigahertz
with our high frequency maps and so

the power of the light on galactic dust
comes from the cross correlation of one

hundred fifty you get Hertz
with three five three it's so

there's not a lot of time left so
I'll get to the bottom line.

So this is the real the.

Distribution of allowed
values of the gravitational.

Sensitive scale a ratio.

That we derive from the joint analysis and
what happened when we

did this analysis which is actually
a complicated bit of analysis was

that instead of a detection with a.

And amplitude of a of about two tenths
what you can infer is an upper limit

with the an amplitude of about the tenth
no detection of gravitational waves their

signal is consistent with being
entirely due to collecting dust.

So so then if we combine that
analysis with the Planck analysis

this is what happens we get an upper
limit on the tents at the scale a ratio

of about point zero nine and
interestingly.

We also can infer that the inflationary
potential is actually very flat.

So we have learned something very
important about the year that that mix of

inflation OK Well there isn't a lot of
time left to reason any time left so

I'll just go.

These results that I've described
the observation resolves a really really

secure OK So and if there's any.

You know question about.

You know whether the results are affected
by instrumental effects or whatever this

in what I've shown you here are the really
secure results from the Planck satellite.

But.

You know we've clearly seen that.

The data is very well described
by a very simple inflationary

model single field inflation model.

And so I think we've closed.

You know a chapter in our understanding
of the very early universe.

But the observations now pose
a number of other questions OK.

Are there actually detectable
gravitational waves for

inflation we've seen that the bicep to
claim you would have been very important

would have given us another.

Pointer to how inflation really works but.

That result is has gone away so
we need to make better observations and

actually try and establish whether
there are gravitational waves.

The the sort of inflationary
models that fit the Planck data.

Have an interesting property that
if the universe ever went through

an inflationary phase like that and
the agreement with life data.

Really strongly suggests that
the universe did then inflationary Ferenc

inflation is eternal.

And so you don't generate
one universe you generate

you know an infinite number
of universes so inflation.

Is eternal.

And so now you know we've solved some
problems you know why the universe is so

old why it's so big where did the
fluctuations come from you know learned

something about the inflationary mechanism
but now we have more fundamental

problems you know if we have our and

eternally in place in the universe then
you know what is the global structure of

the multiverse of this multiverse we don't
have the mathematics to describe that.

OK so that's a really fundamental problem
does inflation require special initial

conditions or do you define what you mean
by initial conditions when you've got

an infinite number of possibilities OK so
that's another really fundamental problem.

And so people have asked you know
are there alternatives to inflation and

I think that you know.

One must always keep an open mind with
these very large extrapolations in energy

scale that may be.

There are other possible explanations so

that's where we've come
you know where we've got.

Cosmologists we've solved some
important basic questions and

it's lead to another layer of even
deeper more difficult questions.

So.

So anyway if you go outside
look at the night sky.

Just imagine we actually do know something
about what happened at the very beginning

of the universe.

So just like to end by saying that
Planck is a big collaboration and

knowledge contributions from scientists
from all of these institutions thank you

thank you for

we open the floor up to questions
I have a quick idea I recognize

several young people in this audience from
our previous event if you're under the age

of eighteen you know youngish and you want
to have some advanced tailor made for

you and you're interested in physics and
astrophysics I'm going to put

my e-mail up on the board I know my
colleagues are looking at me like.

And you can email me and I'll see if we
can create an event for you OK something

where you get to meet the Speaker maybe or
we tailor some topic you like so

I want to do it now before everyone ran
away because you have to go to bed so

you're going to put my name up there you
can e-mail me and say my parents that I'm

allowed if you want to get involve
with more physics activities.

So we'll open up for questions.

And so on so.

I was wondering if you think there
are any ground based larger.

Projects in the near future that may
be able to place better constraints

on the tensor scale or ratio.

Right so.

So it's interesting that there
was a successful balloon flight.

From.

From Antarctica this is indeed
an experiment called spider and so

that experiment works
works very very well.

And so.

So you know I think that you know.

That experiment will get.

Results I mean this should get results in
the public domain you know within a year

or so is what I would expect and
also the biceps themselves.

Have been running in the US experiment.

So they've been running.

One hundred gigahertz.

Receiver.

And they have a lot of data one hundred
gigahertz which they're processing

at the moment so that's different from
the one hundred fifty gig you heard stated

that we analyzed jointly So
the more frequency coverage you have

the more discriminator e
power you have for dust but

I think that the it's you know the
detection is not going to happen you know

I think it's not going to happen
anytime soon because if there is a real

gravitational wave signal with an
amplitude of you know point zero five or

ten scale ratio point five
then you need higher precision

on the foreground monitoring
on the dust monitoring.

Than the experiments would actually have
because you know you'd want to be able to

do this and show a residual as you
know some signal at several standard

deviations I don't think these you know
either the bicep team or the spy the team

will have that level of significance in
that you know in the next couple of years.

Well several questions
were asked one first.

How compression does to be
during different units.

How can pressure and density be equal
that there are different units because

because I'm I'm
a theoretical physicist and

I can choose whatever units I want OK And
so I chose the speed of light

to be equal to unity OK so if you want to.

In high school units is minus Rossi

school.

Yeah.

This is to both of you are really
anybody if I wanted to.

Give an introduction to
this information to.

Middle schoolers or freshman you
know in high school something.

Do you know of any sources that
might be able to break it down for

them a little bit that
I could bring to them.

That wouldn't like to have
their eyes glaze over and.

I want to you know it's exciting
learning is exciting but for

some of them you know if it's if they
don't get it within the first five

seconds they're like you know.

That I mean.

I.
Still find it difficult to think

physically was minus wrong really me OK So
I think.

You know to.

You know to high school kids
probably the best you can do.

You know.

Like in my in my little cartoon.

You know it just it just suddenly.

It's really fast.

And it happens because the universe
is full of really funny stuff

OK I don't know.

It's true so
I don't have a better answer Lapidus but

it is curious that one book I liked I
kept on called black holes time warps and

Einstein's outrageous
legacy that I think is so

Kip Thorne Blackhorse time arps and
Einstein's outrageous legacy I think

that at least is a good read that you
know might be a more appropriate but

that's a black horse not because
mass which is harder I think you.

I do appreciate.

My questions for right.

I think I need that book as well but I
didn't I didn't get how you went from four

hundred years and years after the birth
of the universe back to turn to

minus thirty five seconds right OK so.

So.

The in general relativity.

Once you generate fluctuations in
special curvature they just froze.

And so it doesn't matter if you see
them four hundred thousand years later

they're absolutely as they were.

At the time they were generated and
so if they were generated

you know before to the minus thirty five
seconds that's how you get this link to

the very early universe.

So you need to you need a bit of G.R.
generous about it to get.

There in a crowd has got.

He's a ringer he's my
graduate student careful.

Both two blank mission what are the other
surprises from see him because

that we should expect in
five to ten years from now.

What surprises or or
other scientific new things yes well.

So I mean the obvious one is it's
gravitational waves and there are you

know several experiments looking for
the pattern from gravitational waves.

Then another thing I mean I've given
you just a you know very small.

You know small number
of results from Planck.

You know that neutrinos have believe
what they are there experimentally have

a mass and we don't know what the mass is.

We can use the Planck data to.

You know tell us how many species
there are OK We get very close to

three you know we found you cannot
have four neutrinos species

that's ruled out by significance love and
we can set limits on the mass.

But if you had our high angular resolution

polarization measurements with the cosmic
microwave background then you should be

able to detect the neutrino masses and
actually detect it and measure its own so

high positions that's another area high
resolution polarization measurements

are a powerful way of looking at
the neutrino sector of the standard model.

So then there's another.

You know interesting area that you know
I I don't know what to make it OK because

you know if we go all the way back
to the Pell spectrum and come on.

That's not doing it.

Then.

And you just see.

I.
Think it may take too long to

fun to go back.

OK So this is the power spectrum.

You see here that this
point is sitting low.

And there's this kind of jiggle here.

OK Now those statistically not really
very significant but they're a bit odd.

OK.

And again if you have very high
sensitivity polarization measurements on

these large angular scales you might
you might see some you know correlated

features which YOU TELL YOU tell you
that there's some new physics so

the number of areas where there will
be new results because of my quick

background.

Besides the cosmic microwave background.

You mentioned I'm very patient always send
them to you know so we know that there

should be like some kind of gravitational
wave bigger around me earlier down

C M B But there is also there should be
something in the middle given by some kind

of neutrino background so that we expect
that the neutrino relays at the after

the Big Bend just constitute some kind of
background that is possible to measure I

mean we know that well there are interact
but yeah I mean that there have been.

You know sort of you know theoretical.

Ideas of how you could measure
directly the neutrino background but

they're very it would be very difficult
to make an experiment do that.

But you know it's conceivable.

Yes Has the point been
able to guide Bice's to

a rather relatively dust free area for
a better.

Number two is there
a twenty night experiment.

From plant.

The next generate there next.

Right yeah yeah.

So our well.

Our three hundred three maps
are now in the public domain so

experiment is can look in your
pick up clean a region of the sky

in the bicep region there
are clean clean regions so so

you know we strip it that data and
experiment is using it.

As for
another satellite projects there's no

approved satellite project beyond plan.

And you know given given
that you know it takes

it takes twenty years to get
one of these projects from.

Your proposal to launch.

You know it's not going
to happen anytime soon.

I'll be even older and fatter.

First our own it is I'm so
excited that you came

all the way from overseas here
now since you asked them to me.

I know you had the courage short but
could you go to the to the.

Strange what time it
was like the last two.

Because I didn't want to get she doubted
that out of their product or presentation.

So it was just.

All right all right all right.

I was just giving you two
schematic examples of where you.

Get an infinite number of universes OK so
one is.

That you know inflation with
the sort of potentials that we

infer from Planck once it
starts it never ends so

you have a you will
continually spawn off universes

you know which may have you know may have
different properties to hour to hour.

So inflation is eternal.

And you know we just sit
in one little pocket.

OK And
there may be another universe over here.

So.

And you know this this global structure
I mean I've put time on this axis here

nobody knows how to time slice
this complicated structure

OK So as a theoretical
construct this is beyond us

OK at the moment it's beyond
the cleverest string theorist OK And

then the second example was from string
theory because you know the String

theory has you know nine or
ten spatial dimensions.

And so the you know the words we only
experience three dimensions the other

dimensions you know have to be small.

And there are many many different
geometrical structures for these.

Hidden Dimensions.

And so you know string theory was designed
to be a unique theory of the physical

world and what people discovered is that
there are many many ways of producing.

You know different geometries which
have different low energy physics.

OK And so that was you know
first thought of as you know

a bad feature but now people have realized
that it's actually a very good feature

because you know you wrap up the
dimensions in one way you get a different

low energy low energy physics you wrap
them up in a different way so that there

is the Hidden Dimensions a kind of like
a D.N.A. And so depending on how you.

You know wrap up these hidden dimensions
you can produce lots of different

universes.

And.

POWERS And you know so some of these many.

Physics like the physics.

Of the Standard Model of particle physics.

But if you have these
different possibilities.

Then there's a quantum
mechanical probability.

Decaying from one to another to another
and again you end up with an infinite

number and in this case an infinite
number of infinite universe.

OK So that is where Because monetary.

You know our understanding
has taken us now that.

You know we have these.

You know apparently.

In a strange constructs of multiverse
instead of one single universe

steeply disturbing.

I find it I find it
deeply disturbing well.

I'm glad you asked that because I
also was curious about the slight but

Aboriginal and the rapid steps
of the survey last question.

How can you know expand faster if
nothing could move faster than my.

Right.

How can the universe expand faster
now I got it I got it I got it.

Yes Well so
you know the principle of rote relativity

is not violated OK fully you know although

you know it doesn't tell you principle of
relativity it doesn't tell you anything

about how fast space can expand
it tells you how fast light or

particles or information can flow.

OK so.

So if you have a situation where space
is expanding really fast OK then.

You you know you have a you end up with a.

With an event horizon.

OK so once something crosses that
event horizon it could never

be in causal contact with
the observer again OK so

the speed of light is still
there is a physical constraint.

So that you know we observe that our
universe is accelerating today OK So

we are actually entering a period of
very low energy inflation now and so

what will happen is that you know galaxies
you know the galaxies that you see from

the Hubble space telescope will move away
from us faster than the speed of light and

we will never see them again never
being closer contact with them again so

we will end up as a super galaxy you know
with no other galaxies around us and

then eventually you know the protons will
decay and the radiation will flow out.

Across the event horizon.

So again that is.

Deeply disturbing.

And that uplifting next thing.