ANNOUNCER: This is a production of Cornell University.
SAUL TEUKOLSKY: My name is Saul Teukolsky I'm the chairman of the Department of Physics. And it's my pleasure to welcome you to this year's Bethe Lectures. For those of you who may need a little bit of refreshing of who is-- who are these lectures named after, their named to honor Hans Bethe, who was a member of the Department of Physics for 70 years.
He came to Cornell in 1935 from Nazi Germany. Even then as a young man, he was already a very famous physicist for his contributions in many areas like quantum physics, nuclear physics, and so on. And it's probably fair to say that he is the most important faculty member that Cornell has ever had. Not only was he responsible for turning a rather backwater Department of Physics into one of the leading departments in the world, but in fact his presence at Cornell helped build up all of the sciences and helped attract many different people to come to join the different departments in Ithaca.
He won the Nobel Prize for his work on explaining how the stars shine, which was an application of nuclear physics. Shortly after doing that work, he went to Los Alamos, where he was the head of the theoretical division of the Manhattan Project that developed the atomic bomb. And after the war, he became one of the leading proponents of arms control and worked very hard in many different ways on the test ban treaties and other related efforts. And it was a combination of his stature as a scientist and his personal integrity that gave him such influence in this area, that even when people disagreed with him, they nevertheless had to take him seriously, and he was probably one of the most influential people in helping shape the policy in this country on this important issue.
He retired in 1975 and then began a whole new second career as an astrophysicist. And it's perhaps fitting that one of the major areas in which he worked during this period was on the theory of supernova explosions. And the reason it's fitting is that tonight's Bethe lecturer is one of the world experts on supernovae from an observational point of view.
And so we're very pleased to have as this year's Bethe Lecturer, Professor Robert Kirshner from Harvard University, where he's a professor of astronomy and also the Clowes professor of science. Bob was an undergraduate at Harvard. He was a PhD student, a graduate student at CalTech in astronomy. And in fact, that's where I first met him. I was a graduate student in physics at the same time. After doing a post-doc at Kitt Peak and then serving on the faculty at the University of Michigan, he went to Harvard in 1986, and he has become one of the leading observational astronomers in the world.
He has received many honors and awards for his work. I'll just mention a few. Last year, he and his collaborators shared the Gruber Prize in cosmology, and some of the work that I think you're going to hear about tonight is related to that. He has been elected a member of the American Academy of Arts and Sciences, the American Philosophical Society, and the National Academy of Sciences. He is also well-known as an expositor of science for the public. And if you enjoy his lecture tonight, I encourage you to take a look at his book the extravagant universe. Please help me welcome Professor Bob Kirshner.
PROFESSOR BOB KIRSHNER: Thank you. Well, it's a great pleasure to be here tonight. What I want to do is to try to describe how the observations of supernovae, these exploding stars, enable us to trace the history of cosmic expansion over the last 13 billion years and how these measurements have given us a surprising result, which is that the expansion of the universe is not slowing down as we expected but speeding up. I show you a picture here of a galaxy, this big fuzzy thing that has 100 billion stars in it, and then this one star down in the corner, which for a little while, for about a month, shines brightly as five billion stars like the sun, those supernova explosions are the key to mapping out this history of cosmic expansion. Now 13 billion years of history in 45 minutes means that I have to go pretty quickly and leave out some of the details, for example, references to other people's work.
So let me start because I'd like to tell this story in a way that-- in which Einstein figures. Well, of course, you couldn't tell it in a way which Einstein did not figure, but in any case it's by thinking about what people knew at the time when Einstein was inventing general relativity, which is the theory of gravity, the thing which applied to the universe gives us the picture for how the universe might change. In 1917 when Einstein was working on these ideas of how gravity worked in the large scale, astronomers were quite confident as we always are about the size and shape of the universe. We were however quite wrong.
The-- in 1917, what people meant by the universe and what astronomers thought the universe consisted of is what today we would call the Milky Way-- the Milky Way galaxy. There are plenty of seats down front. Don't be shy come in. Because if you come in and sit down, you can't leave so easily. Come in Come in.
In 1917, the universe-- what people thought was the universe was the sister of stars, the Milky Way, the system of which the sun is a member. Today we know-- and I'll describe a little bit how this change came about-- today we know that our Milky Way galaxy is only one in 100 billion galaxies like the Milky Way. And so the universe is a much bigger place. The problem we have to explain is a much broader one than the one that Einstein was focusing on in 1917.
One thing I want to emphasize is how much the advances in technology have made a difference to our ability to understand the universe. And I'll start with this picture of Einstein, and you can see, for example, that this was a photograph taken no doubt with a photographic plate and that the photographic plate was a pretty slow tool for measuring the light. Because during the time that Einstein was sitting here, you notice that this person got up and walked away and this person was shaking his head violently in disagreement with Einstein.
The other thing that you may not realize is that we now have the ability to see what people were thinking at the time when these photographs were taken. And what Einstein was thinking about the universe at that time was that it must be static, that is this universe, by which he meant the Milky Way, which we know is not today's answer, must be static, not expanding or contracting. At least he thought that might be the right answer. And when he looked at his equations, he found that to a way to do that that was satisfactory to him at that moment was to put in by hand an extra term for which there was no real justification from the physics to make a static universe. So here's what he said. Can you read this from the back?
No? Well, it's a German. It says that term is necessary only for the purpose of making possible a quasi static distribution of matter as required by the fact of the small velocities of the stars. So what Einstein was talking about was the thing that was static were the Milky Way galaxy. That's what they thought was the universe. And the things that he was looking at to see whether this was true or not were the stars.
This picture of Milky Way-- the Milky Way as the universe and the stars as the objects to look at was something which changed very rapidly. But that was Einstein's motivation back in 1917, to make a static universe. Now how did this idea of a static universe and a small one get broken?
It happened quite quickly and it happened partly because of the scientific work of this modest woman, Henrietta Swann Leavitt, who worked at the Harvard College Observatory. She worked on a great stack of photographic plates that came from the observatory in Peru that Harvard had been running, and she was given the assignment of studying variable stars a particular kind of variable stars. And she found something very interesting.
In a set of these stars that are all at the same distance, there was a relation between how rapidly they varied and bright they were. And the reason why this was important is something like how rapidly the star varies, how long it takes to get bright and dim and bright again, is something that doesn't depend on the distance. And so if you can measure that, you can tell whether you're looking at something that's the equivalent of a 100- watt bulb or a 75-watt bulb or a 50-watt bulb, you can tell what the intrinsic brightness of the source is. And then by measuring how bright the star appears, you could figure out how far away the object is that has that kind of star in it.
And this idea of looking at stars or objects in distant places and figuring out how far away they are by how bright they appear is a standard trick in astronomy. But again this became much more powerful in these years right after World War I by the introduction of new large telescopes. So here I show you the 100-inch telescope-- of course, now it's 2 and 1/2 meters at Mount Wilson which is up above the little city of Los Angeles out in Southern California. And at the time when the telescope was built, the population of Los Angeles was about 25,000.
And the city was quite dark down low at night. The site is still good. The telescopes still works, but it is true that the Los Angeles basin has gotten a little brighter. And the most difficult work is not being done with the Mount Wilson telescope.
At any rate, Edwin Hubble, who was working with this telescope at Mount Wilson, used photographic plates of the type I've described. In fact, here's the photographic plate holder. The plate goes right in there, big glass sheet with goo on it that detects the incoming light. And he looked for the kinds of stars that Henrietta Leavitt had discovered.
And by doing this-- by finding stars like that in the nebula-- these fuzzy little patches, these spiral things that you may know and I'll show you some in the second-- he was able to show in the 1920s that we live in a big universe, not a small one and that the Milky Way is not the whole universe. It's just one in a whole galaxy-- in a whole universe of galaxies like ours.
So here's how he did it. Here's a picture of the Andromeda nebula M31 as we call it. And by taking photographs with that telescope that I showed you a second ago over and over, he was able to find the variable stars, see how long their periods were. And by seeing that they were very much fainter than the ones that Henrietta Leavitt had measured in nearby systems, he was able to see that the distance to this galaxy was very large.
Now one of the hard-- distances in astronomy are always the hardest thing to measure and the hardest thing to get your mind around. The reason that it's important though is that the information that we get from astronomical objects comes to us in the form of light. And light, although it travels very fast, it doesn't travel infinitely fast. It travels a foot-- that's a unit of distance that's used in the United States and in Myanmar-- a foot in a nanosecond-- in a billionth of a second.
So when I look out in the room, I never see you the way you are. I see you in the words of the song the way you were. So when I look at the people in the front, I see them as they were maybe 15 nanoseconds ago. Curiously the people in the back looked younger because I see them the way they looked 100 nanoseconds ago. Just a joke in the room but not a joke in the universe.
In the universe, when you look at nearby things, the stars in our own Milky Way galaxy, you're looking at light that has traveled to us for years or tens of years or hundreds of years, say since the Federalists controlled both houses of Congress. Anyway when you look at nearby galaxies, though, you're looking at distances that correspond to millions of years, and so they are millions of light years away.
What I'm going to be talking about tonight though is distances that are 1,000 times bigger than that, billions of light years, 1,000 millions. And by using stars which are much brighter than the ones that Henrietta Leavitt studied, we can measure the distances to those from the supernovae. And that's how we plot out what has been happening in the past. Because the light comes to us from the past, a telescope is a kind of no nonsense time machine that lets you see what the universe was like in the past compared to what's happening now. And from doing those measurements, we can see whether the universe is speeding up or slowing down as I'll show you. But I had better speed up.
So the universe is-- we think now-- a way to think about it is not as that one island of stars but that there are many such places. And if you take a picture with the Hubble Space Telescope-- it's named after Edwin Hubble-- it's about the same size as the 100-inch telescope at Mount Wilson, but it's at a much better site. And it allows you above the Earth's atmosphere to take a picture of what any part of the sky. This is a blank part of the sky to ordinary photographs. And when you take a long exposures so that you accumulate lots of light by pointing at this spot for many hours, you find that it's full of galaxies.
So here are some that look something like the Andromeda Galaxy I showed you a minute ago, big whirling spirals. But almost everything you see in this picture is a galaxy, even these little things that look like little shreds are themselves giant systems of stars, 100 billion stars, but just very far away. So instead of looking at the properties of the stars, of course, looking at the properties of the galaxies might be the way to find out about what the universe is doing in the large scale.
One thing we know is that our own sun is in the Milky Way, but we haven't made the mistake-- well, we did make the mistake of thinking that the Earth was at the center of things-- but one of the other things that was discovered in the same years as the fact that the Milky Way was not alone as an island universe or as a separate galaxy is that we're not at the center of the Milky Way. So our own location, our own sun, is located out in the spiral arms of a galaxy somewhat like this one. In fact, Ithaca is not exactly at the center of the galaxy although it's certainly at the center of your own world. Of course, I didn't mean anything like that. But we were out in the sun and the earth and New York state are located out far from the center of our own Milky Way galaxy.
And the reason I bring that up is of course, this idea that we're not at the center is something which we have slowly learned through the history of science. Now if Hubble had done nothing more than show that we live in a big universe in which the Milky Way is just one of a large number of galaxies, you would be a very famous astronomer. But it did something else, which is in the years just before Hubble did its work on the distances, Vesto Slipher and others took the light from a galaxy and spread it out into a rainbow into a spectrum.
And by looking at the properties of this spectrum, you can tell something about what the temperature of a star is. You can tell something about the chemical elements that are in the atmosphere of the star. And you can also tell something about its motion. So I show you here-- here's what astronomer does is take something beautiful like a rainbow and turn it into a graph up here.
And here you'll see these dark lines that cross the spectrum, or the dips in this graph that represent how much light there is at each wavelength, have some very distinct patterns in them, which are set by the properties of the atoms in the atmosphere of the stars. So calcium we know makes absorption here, and sodium makes an absorption here. The chemistry of the sun and the chemistry of other stars is something that we can find out about.
The other thing that you can do from measuring these lines is to measure their wavelengths. And what people found was that the spectra of galaxies showed systematic stretching out toward the red or in a few cases compression toward the blue of this spectrum. And we know what causes that. Motion causes that.
The objects that are moving away from us get there spectrum stretched out to the red. That's called the redshift, or locally it's called the big redshift I believe. And when things are coming toward you, they are-- the spectrum is crunched down to the blue. So that means we can measure whether a galaxy is coming towards us or away from us.
And Hubble-- so Hubble had these two lists of numbers. He had a list of distances from how bright those variable stars were and a list of velocities from how stretched out the spectrum was. And here's a plot. This is its own diagram from 1929. He didn't call it the Hubble diagram. He called it the diagram.
This was velocity going this way and distance going that way. And so for each galaxy for which he had measured a velocity and distance, he plotted them on here. Here's 0 right here. So here are nearby galaxies where the distance is small. Some of them are positive, that is going away from us, redshifted. But a few are blueshifted in fact.
But as you go farther out what you notice is the preponderance-- well, all of them-- are redshifted, and as you go even farther out, they're redshifted more. This was a very surprising and interesting thing that showed that the galaxies that are near to us are moving away from us. The ones that are distant, moving away more rapidly. This is what we call Hubble's law that the redshift is proportional to the distance, that there is a slope of this line so that as you look at more and more distant objects, they're moving away from us more and more rapidly.
Now I took pains a few minutes ago to make an argument that we're not at the center of things because I think this kind of diagram makes people think well maybe we are at the center of things. At my institution, most of the undergraduates and frankly all of the faculty believe that this is the basic picture that describes how things are organized. It's what I've called the egocentric universe, and so it has me at the center and then all of these galaxies and all the rest of the universe moving away from us.
This seems a little self-important if I may say. And yet you might wonder how could it be that everything's moving away from us, and we're not at the center. Well, it turns out that if the universe is stretching out in all directions, then it's not necessary for us to be at the center. In fact, any galaxy would have the same view.
Now I've said that, and you know I was moving my hands a lot. That's always a bad sign, but I'm going to use the miracles of animation to try to convince you that this is true. And so here we have a set or-- a picture, and all I've done is stretch it out in all directions. I just enlarged it. I didn't do anything else.
And you notice that when this set of pictures is lined up around that galaxy that it looks like everything's moving away from that one. And in fact, if you look at the ones that are nearby, they're moving a little bit. The ones that are farther away seem to be moving faster. So something that is just enlarged in all directions stretched out equally in all directions looks to you like everyone's moving away from you.
OK you may say, but didn't you pick a particular galaxy? Well, yes I did. But let's try another one. Let's try that one-- oh, that one right there.
So now this is the same set of pictures. The only thing I've done is to look at it from the point of view of that galaxy. And if all goes well-- so here's the same set of pictures. It's just stretched out in all directions. There was no center to it. And if you line things up on this galaxy, it looks like everything's moving away from it. The nearby things moving away slowly, the more distant ones moving away rapidly.
So Hubble's law, this discovery that we live in a universe in which things are moving away from us, we don't think it's something special to us. We think that that is the typical view in the universe. And what that really means is that the whole universe is stretching out in all directions. It's an expanding universe.
Now you may remember, I went to a lot of trouble at the beginning to explain why Einstein invented a static universe. And eventually he had to go talk to the observers, and here he is visiting the observatory, the Mount Wilson Observatory offices at 813 Santa Barbara Street in Pasadena, California. Here's Einstein of course. He's been writing some things here on the blackboard, still holding the chalk I notice. Here's the observatory director, very proud. Einstein is here excellent.
This is Hubble over here. This is George Ellery Hale. George Ellery Hale is the man who built the 100-inch telescope and later the 200-inch telescope, and here he is patting Hubble on the head. Very good.
So at this point, Einstein really gave up this idea that the cosmological constant was making a static universe. He never liked it anyway, he said. It's mathematically unsatisfactory. Sorry I did it. And Einstein for the most part stopped talking about the cosmological constant, something that pushes back against gravity.
Why do we call it Einstein's greatest blunder you might ask. Well, it seems like a blunder all right, but the real reason we call it Einstein's greatest blunder is that George Gamow told us that it was his biggest blunder. Here this is from Gamow's autobiography. As we say no one else could have written it.
And he says, "Einstein's original gravity "-- that's without the cosmological constant-- "was correct and changing it was a mistake. Much later when I was discussing cosmological problems with Einstein"-- I wanted to put that in my book, but no such luck-- "much later when I was discussing cosmological problems with Einstein, he remarked that the introduction of the cosmological term"-- somehow that line should be one line down-- "was the biggest blunder he ever made in his life. But this blunder rejected by Einstein is still sometimes used by cosmologists even today, and the cosmological constant denoted by the Greek letter lambda rears its ugly head again and again and again."
Gamow, very good. Very good. Well, I'm here tonight to tell you that the evidence much, much later is that the universe is not just expanding but expanding faster over time and that you need something to do that and the something may turn out to be very much like Einstein's cosmological constant. That's what I meant by saying the blunder undone. And that while it may have reared its ugly head in this case, there's really evidence that you need something very much like that to explain the facts. So let me get to that story.
Einstein gave up on this and said let's not talk about this cosmological constant anymore. In fact though the-- his associates at about 1930, after the discovery of the expanding universe, were not so quick to give up this cosmological constant. De sitter, who was a Dutch astronomer who read Einstein's papers and understood them, was not so sure that the expansion that Hubble was seeing wasn't the result of the cosmological constant pushing and making things go. So here you see a Dutch cartoon in the shape of the letter Greek letter lambda.
This is Professor de Sitter, and he's making things blow up. Here it is in Dutch. Dutch is-- English is a dialect of Frisian, so you can read this-- who however blows up the ball-- the ball up, get it. What makes the universe expand or swell up, upswelt? That's got to be lambda. Any other answer you can't give. So it isn't true that people gave up instantly on the cosmological constant, and what's true now is that it looks like we may have to come back to something like it to explain the observational facts.
Of course, if you go to the National Academy and you go to the big statue of Einstein on the mall and you look at the basic equation for general relativity up here-- could be in one of your books-- there is no lambda. There's no lambda in it. He also has a couple of other equations you might be interested in. Here's the photoelectric effect. That's good. Oh, give him the Nobel Prize for that. And here's e equals mc squared, the equation that describes how transformations of nuclei might yield energy, a problem which haunts Bethe, of course, paid detailed attention to in the 1930s and 40s.
In fact, here he is looking at this. And so Bethe's contribution-- many, many contributions-- but one of them was to go beyond the idea that you could get the energy out by changing hydrogen into helium to show us how that happened in stars. And here's a picture that I think might be from the earlier time when he first came to Cornell at least judging by the cars. He seems the same, but the cars seem to have gotten quite old.
So how do you make a test of the history of cosmic expansion, whether things have been speeding up due to a cosmological constant or slowing down due to gravity? Here's a picture. This is de Sitter. This is Einstein. This is Einstein de Sitter's space.
And what you can find here is a diagram that shows the relation between velocity and distance, a Hubble diagram. And the idea is that you should look at large distances, look into the past, and compare the light that you get from the distant past to the light that you get from nearby from the nearby galaxies so the nearby supernovae. So that's the idea, and I'll show you how we've gone about doing that.
Now Hubble noticed something else which turned out to be very important for this story, which is that in the nebulae-- so here's one of these spiral nebulae-- from time to time, they were rare but quite conspicuous. There were stars that flared up. Now we know about stars that flare up in our own Milky Way galaxy-- and they're called novae-- but this is-- of course, if Hubble was right that the distances were very large, then these novae had to be extra bright. They had to be super novae. Hubble was on to it. He said there's a mysterious class of exceptional novae, which attain luminosities that are respectable fractions of the total luminosities of the systems in which they appear.
He wasn't a real quantitative guy, respectable fractions. Anyway-- [INAUDIBLE] fractions. And Fritz Zwicky, who worked at CalTech, took this idea seriously and tried to find out what those supernovae were. So here's Fritz. This is famous picture of Fritz taken much later in his life. Here he is demonstrating his opinion of the other people in the department.
This gesture he referred to as the spherical bastard. He said, "They're bastards any way you look at them." But here he is in a more constructive mode operating the first Schmidt telescope at Palomar, the little 18-inch Schmidt that he had commissioned for this purpose. Took in a big chunk of the sky so he could search for lots of galaxies for these very rare phenomenon. You need to look at a lot of galaxies because supernovae, although they're very bright, are very uncommon.
I mentioned that they're brighter than billions of stars like the sun. They are a million times brighter than the stars that Henrietta Leavitt had studied that led to opening up the measurement to other galaxies, but they're rare. There's about one supernova per century in a galaxy. So if you want to study these things, you have to think of some way to look at lots of galaxies. And Zwicky really is the one who pioneered that.
Of course, it's a big universe. There are about 30 per second going off in the universe, so the fact that we only find a few hundred each year just shows how lazy we are. We really should do much better.
We think we understand what the supernovae are now having studied many of them and that they are white dwarfs, carbon and oxygen nuggets leftover after the burning of a star, in which a neighbor star, a binary star that is these two are orbiting around one another, puts mass onto this otherwise stable star and brings it to a position where it will explode as a thermonuclear bomb. The carbon and oxygen that are the ashes from a low mass star are stable unless they are provoked.
But if a neighbor star puts enough mass on, you can get to the point where that material begins to burn. And it burns suddenly, violently to make a tremendous explosion that we can see halfway across the universe. What's more, there is a limit to the mass of white dwarfs, about 1.4 times the mass of the sun, so that you might expect that the explosions of these white wharfs would be more or less similar anyway limited by the fact that they have somewhat similar masses.
These objects are rare, and there's nobody in the room who has seen a supernova in our own galaxy. Even Hans Bethe did not see a supernova in our own galaxy, but Tycho did. Here's a picture of Tycho, the Danish astronomer who was very young-- and it's not a picture of him in 1572-- but noticed a new star in the sky brighter than any of the others, brightest of the planets. And that could happen any day now. We could have another supernova in our galaxy if only we had an astronomer on earth who was as good as Tycho. What am I talking about?
If you look at Tycho's supernova today-- this is an image taken with an X-ray telescope at the site where the explosion went off a little over 400 years ago-- you see stuff rushing out. And in fact we can tell what the chemistry of this material is. It's mostly iron, which is what you expect to get when you take carbon and oxygen and burn it as much as you can by nuclear fusion.
So the supernovae are very important in that way. They make the chemical elements. So, for example, if you want to see something where-- whose elements were made in a supernova-- here's something-- you don't think this is made in Detroit? Well, yes, it was, but the iron that's in it, the nuclei, were made in supernovae that were-- that took place before the earth formed. That was a great success. Let's keep moving.
Here's another group of things that were made in supernovae. Since the atoms in the iron in your blood or the calcium in your bones were made in supernovae, you are actually the product of supernova explosions. You are star material. And this is my research group.
One thing we've been doing is using our telescopes in Arizona. These are fairly modest-sized ones, but where we can get the observations when we need them of supernovae as they explode. And what we measure is how bright the supernovae are. One thing that is interesting is that the supernova explosion gets bright and dim in about a month.
So the scale down here-- here's 20 days, 40 days-- so the time scale for these things is a month. That's about the time between one new moon and the next. So the natural rhythm for searching for supernovae is to observe each month, compare the pictures from last month to the ones from this month, and to find the new objects.
So here's a diagram, which is like Hubble's diagram, only now for supernovae, where the velocity is plotted up this way and the distance out here nearby-- these are relatively nearby. This is only two billion light years on the horizontal axis. Nobody laughed, so I've succeeded in convincing you that two billion is not such a big number. It's like talking about the economy. Anyway, here's two billion light years on the scale.
Hubble's Hubble diagram fit in this little red square down here at the bottom. So Hubble's original work was definitely on the right path. The new tools that we have now including much brighter standard candles using the supernovae allow us to make these measurements to larger distances, but nearby the relation that the velocity is proportional to the distance seems to hold.
So the gain is to look nearby to find expansion and far away to see whether the expansion in the past was faster or slower. If the universe has been slowing down, then the expansion back then would have been faster. If the universe has been speeding up-- yes, has been speeding up, then the expansion back then would have been slower. So the idea is to compare the expansion rate back then to the expansion rate here. That means you have to find very distant supernovae.
We know that there's something out there that will make things slow down, and that's gravitation. There's gravity that's associated with all the galaxies, all the stars, and the matter in between the stars, the so-called dark matter. We know that it's there because we see it at work. We see the presence of dark matter bending the light from distant objects behind a cluster of galaxies.
So here these yellow things are all galaxies, many at the same distance. Objects that are farther behind actually have their light distorted into these arcs. Maybe you can see them. These arc-like features, which I would point out to you if I hadn't exhausted the battery earlier, that show you that there's-- the light is bent-- the light from objects behind is bent as it passes through this kind of gravitational lens.
We also know that there's mass there because the galaxies in these clusters are moving around very rapidly. We also know that there is mass there because there's very, very hot gas in these clusters of galaxies, that it's sitting in the strong gravitational pull of a lot more mass than is associated with the stars themselves. So there's mass in the universe, and it should be slowing the universe down.
So the idea is to look into the distant past and to see if that's really true. And that's what we thought we were going to do when we started on this about 10 years ago. The big step forward comes from electronic detectors, from better detectors that allow you to take a picture of a big chunk of the sky and find the supernovae in it.
And so here's a picture of a chunk of the sky with many thousands of dots in it. Each of those dots is a galaxy. And if you were attentive enough to compare an old one to a new one, you'd be able to-- thank you. If you're attentive enough to compare an old picture to a new picture to every galaxy in the picture, you can look for things that have exploded.
People won't do that. Even graduate students won't do that. It turns out you really have to train a computer to look at many thousands of galaxies for this sort of thing. And so here's Brian Schmidt, 10 years ago, explaining to me how easy this is going to be. And that there was another team working on it. They've been working on the software for three years. I said, oh, Brian, this going to be really hard. He said, yeah, it is hard. It'll take me a month.
So here's what he did, and here's the idea. This is 1/1,000 of the image area of one of those big detectors. And here's some galaxies and stuff in it. The idea is you do a subtraction and things that don't change go away, and the things that do change like a new star erupting in that galaxy show up as a dot.
So the technology for finding these things comes from big detectors, which are fabricated in the same way computer chips are, and from them computers themselves that will do the-- help us do this intensive work of comparing one sample to another. So here is the crew that I was working with in 1996 I think. So here's Brian Schmidt, who had been my graduate student, and Adam Riess, who was my graduate student, and Sorb Jha, who was my graduate-- there's John [? Tonery. ?] Here's Bruno [? Lidengood, ?] who was a post-doc. Anyway a bunch of us were working on this stuff.
And we had measurements of the brightness so we could figure out the distance and the redshift for a handful of very distant supernovae whose light had been traveling to us from about halfway back to the Big Bang. And here's Adam's notebook, which he very kindly gave to me. He said, you should show this. And it shows his crummy handwriting.
But what he worked out-- what the maths of the universe would be in this funny language. Here's omega. That's a Greek letter, the last letter of the Greek alphabet. It tells what fraction of the universe would be in massive stuff. And if you worked out for a universe where there was no cosmological constant, omega lambda with zero, he got a number that was minus 0.36.
Minus. That seemed very strange that you had to have negative mass. And what it meant was that this was not the right way to do the analysis. That, in fact, you should take into account the accel-- the possible acceleration of the universe.
So we were puzzled by this for a while but not for too long. And by the beginning of 1998, we were able to conclude from a fairly small sample that the universe appeared to be accelerating. The universe will be a very lonely place to look at says, Professor Robert Kirshner. Well, that may not turn out to be true.
Anyway, it was a big deal, and it surprised Einstein quite a lot as you can see here. Of course, he'd been dead for several decades, but in any case, there are two teams that work. The-- our High-Z Team and Saul Perlmutter's group, the Supernova Cosmology Project, both of which found this bizarre thing that the universe is speeding up.
Here's what the data looked like today. If you have a lot of-- so this is a Hubble diagram. The coordinates are a little bit different. But anyway they're low redshift supernovae down here and high redshift supernovae down here. And if you look in this graph down here, if the universe were slowing down due to gravity in the way many people thought it ought to, the points would lie along this red dash line down here. Here are the nearby ones. Here the distant ones.
And you can see that that's not right. The distant ones all are lying above that point. What that means is that during the time while the light was traveling from a distant supernova to us, the supernova-- the galaxy-- the something-- the universe, that's it. I knew it-- the universe expanded more than you would have expected otherwise. The distance was larger, and as a result, the supernova appeared a little dimmer. So that's the evidence the difference between where these points lie and these lines down here.
It's a subtle difference, but it's a real difference. And we're quite sure now after a decade's work that this measurement is not mistaken, that we really do see that the universe has stretched out an extra bit in the recent half. We're looking back about five billion light years to a time when the universe was about a third of its current age. And it's so certain that now you can get a license plate issued by the state of Maine that says XLR8N universe.
There are many lines of evidence, though, that seem to be converging. I think people would not believe it if it were just the supernovae pointing to this incredible result. And this diagram, which is a little messy, shows you how much of the universe is in the form of the dark energy-- so that's up here-- and how much of the universe is in the form of dark matter, and that's shown out here. The supernova results themselves restrain things to this zone in the diagram, but there are other measurements of how galaxies cluster and of the fluctuations in the glow from the Big Bang itself that we see in all directions that converge on a single point here with this x, which is a universe that is about one third dark matter and about one-- about two thirds dark energy.
So it's a very funny new picture that we have. And here's a diagram that indicates the contents of the universe. The gravitating stuff is in this picture about 27%, and the dark energy is about 73%.
Now my theoretical colleagues are very proud of this diagram. They say look, it is full. And I just caution all of you that when someone shows you something like that, you should ask what is it full of? And in this case, it is full of ignorance, which is a good thing. It means there's plenty more for us to do, that we don't know what the dark energy is. It might be Einstein's cosmological constant, but it might be some other thing that acts like the cosmological constant but isn't really the same.
The matter over here is also quite strange. You notice that there's a wedge here that refers to atoms, so that would be the atoms in stars or in the gas between stars and then other stuff that's called dark matter but which is not made of atoms. So we think that this stuff is real because we measure its effect on the motions of galaxies, for example. But if we think about what conditions were like in the early universe, the cooking of elements in the Big Bang itself makes us think that this stuff can't be neutrons and protons, the ordinary stuff that we're made of.
So I don't know how that makes you feel, that you're in this little 4%. Personally, I feel quite special myself. But it does mean that the stars and galaxies that we observe with our telescopes are not most of the universe. They show us what's happening. There's a kind of tug of war going on between the dark matter trying to slow things down, the dark energy trying to speed things up, but the stars and galaxies themselves are just tracing what's happening. They are not most of the stuff of the universe.
It's a little like looking out on a mountain range on a moonlit night, and you see snow up on the ridge. And you see-- you know where the mountains are, but the snow is not the mountain. The mountain is something else. The snow is just showing you where that material is.
And in the same way, the things that we can see are not themselves, most of the universe. They are giving us a clue to this curious struggle that's going on between the dark matter trying to slow things down and the dark energy trying to speed things up. At the moment, although we have some ideas about what each of those things is, we don't really know what they are.
Now that doesn't make a whole lot of sense. And I found that out when I went to an institution in another country, and they had a gift shop. And I went in there because it was interesting, and they had t-shirts with a smart aleck sayings on them that you could buy including one that turned out I had said.
It said, oh, the universe is under no obligation to make sense. Your PhD thesis is. So I thought that was pretty good. And when I said, gee, these shirts have something that I said, they said we'll give you a good discount.
So the dark energy is real, but we don't really know what it is. If you Google dark energy, it used to be that you got this, which is from American Hydroponics, a company that makes plant food, so that you can grow controlled substances in your closet. And what they say is something that's editorialized. They editorialize quite a bit. They say these processes are responsible for the very distinct odor of dark energy.
And it is true that there is a kind of odor to the dark energy. If it's the cosmological constant, that was Einstein's greatest blunder-- well, Gamow said so-- there's been a slightly disreputable idea, but it looks like we're forced to it. What's more there is a theoretical way to estimate what the vacuum ought to do, what empty space ought to do and how gravity ought to operate in the empty space. I am authorized to speak on this because I have a union card from the International Brotherhood of Theorists. I only point out that it's valid to infinity. They can't take it away from me.
And if you take an envelope and do a little calculation on the back about what you expect for the energy density of the vacuum due to gravity. It's a very large number. It's about 10 to the 120 in units where the observed value is 0.7. This is what in academic circles we politely call not good quantitative agreement. Try it on your exam.
Anyway it's what Steven Weinberg more vividly has called a bone in the throat of physics. So there's a really big problem here, which is that the understanding of how gravity works on the small scale and in the vacuum is something which we haven't connected up with quantum mechanics in the proper way.
When you make a first stab at it doing this, you get an answer which is quantitatively way out of whack. So there's something missing. There's something more that needs to be done, and it may be that one of you will be the one to do it.
One thing you can do is you can think about the cosmological constant as being something that is a source of gravitation over on the right hand side of Einstein's equations. So I thought I would try to do that until the park police came and told me to stop it.
Now you might say, well, this all fits together. You've explained this story. It's kind of a mystery, but are there any predictions that this model makes that we can test? And one of the ideas is that if you have a mixed dark energy and dark matter world, then as time goes by, the dark energy-- if it's the cosmological constant-- will stay the same whereas the density of matter in an expanding universe will go down. So the balance between dark energy, which tries to speed things up, and dark matter, which tries to slow things down, will shift over time.
The trouble is we can't see into the future, but I described how we can see into the past, which is using telescopes. So here's a picture of the Hubble Space Telescope, which we've used to look into the distant past to see whether this process of the dark energy staying the same and that dark matter getting diluted is really taking place. So the idea is if you look far enough back, the dark matter should have had the upper hand. The universe should have been decelerating before it was accelerating.
And interestingly enough, you can make these measurements. Adam Reiss-- who I showed you before-- led the work on this. Here's the detector for this, which astronauts brought up in the shuttle and replaced the-- to replace the instrument in 2002 and which we've been using ever since although it actually it broke last year. Well, we've been using until last year.
Astronauts are scheduled-- it says October 2008. Sorry, I should have fixed the slide-- that got canceled. The trip that was supposed to be last week got canceled, but it's been pushed forward into next year. But there will be a chance next year for astronauts to go up and repair the things that are broken on the Hubble Space Telescope.
In any case, we were able to use that instrument to find distant supernovae. And let me show you this. Well, this is a supernova in a distant galaxy. That cost a million dollars. Let's do it again. There we go.
There's the supernova in a very distant galaxy, and here is a plot of the brightness of the supernova. It looks a lot like the ones that we measured from our little telescope in Arizona except the scale shows that it's very much fainter because it's very much farther away. And by working out how far away and how fast those supernovae were going, we can test this idea that the dark energy was smaller-- who had a smaller contribution, that the dark matter was more important in the past. And it turns out that it really is, that the universe was slowing down before it speeded up.
Now those of you who have studied physics know that the change in position is velocity. The change in velocity is acceleration. And to your undying amusement, the change in acceleration is called the jerk. This means that the universe has changed from decelerating to accelerating, so there's a point where the universe had jerk.
And here under-- on The New York Times under a headline that says "A Cosmic Jerk" is a picture of Adam Riess. It's really fantastic. Even his mother doesn't like this. In any case, we've seen the evidence for this change this thing that you might predict.
What about the future? Well, I said the telescopes are not so good for seeing into the future. We don't know exactly what will happen. It depends on the properties of the dark energy, which we haven't determined. It could be that there'll be acceleration indefinitely into the future in which case that guy talking about it being a lonely, dark, cold place could be right.
It could be though, depending on the properties of the dark energy, that other things happen that there could even be a collapse in the future or of a much more violent expansion that would rip things apart. We're really interested in this because it is a big problem in physics. It lives right at the heart of understanding what gravity is, and so we have an idea that maybe it would be worthwhile to make these measurements more accurately and over a bigger span. And so there's a whole effort right now to work on a specialized Space Telescope, which would be just for the purpose of finding out about the dark energy, and we'll see whether that comes to pass.
Well, that's enough. I promised to go on about the 13 billion years of the universe, and maybe now it seems a little bit as if I really have. I just wanted to say a couple of things. I spent some time as a president of the American Astronomical Society. When you're president of a learned society like that, every once in a while you're called upon to explain to people-- well, actually congressmen-- why it is that you need these vast sums of money to investigate these areas of science.
And the answers that are given often by heads of physics societies or chemistry societies or any of these organizations usually go in categories like this, that you say something like, well, it's very important to do fundamental science because it leads to advances in technology. Or it's very important to have the upper hand in dark energy because it's a dangerous world out there, and this could possibly be important in defense. Or since a lot of the congressmen are old guys, it's-- you often say something about, well, this is very important to human health and so on.
And I am not opposed to any of those things. I think being rich and safe and more or less immortal would not be such a bad outcome. On the other hand, I don't think that's the reason why we do this sort of work, and I don't think it's a sufficient reason either. It seems to me that if those things were true but you are bored that it really wouldn't be such a great world and that people are really curious. We want to know where we came from. We want to know what the universe is made of. And we're embarked on a tremendous adventure to do that. So what I would say, and what I did say-- although not to great applause to that audience, I must say-- is that we're really doing this for the joy of finding out how the world works. Thank you very much.
SAUL TEUKOLSKY: Questions?
AUDIENCE: Thank you for helping me accelerate away from boredom for the last--
PROFESSOR BOB KIRSHNER: Yes. Yes, good.
AUDIENCE: I have a very naive question. So if a redshift is providing us with evidence for the accelerated universe, how do you collect the blueshift?
PROFESSOR BOB KIRSHNER: Yeah. Yeah. There are some galaxies which are very nearby for which the gravitational attraction of our galaxy for that other one is sufficient so that they are actually moving together under the force of gravity. So in addition to this overall expansion, there are also local motions that have to do with the density in homogeneous-- the lumpy, dense places-- the low density places. The dense places can pull things in. The low density places lose stuff. So in addition to the overall expansion, there can be additional motions in the universe.
AUDIENCE: [INAUDIBLE] Both are very close to [INAUDIBLE].
PROFESSOR BOB KIRSHNER: Yes. That's right. An M31, for example, is very close to us, only two million light years away. It's coming toward us, and you should check your homeowners insurance to see if you're covered for galactic collisions.
No, well, sure. Go ahead.
AUDIENCE: How far away have we been able to observe anything yet or how far back?
PROFESSOR BOB KIRSHNER: Yeah. So these supernovae that I'm talking about take you two thirds of the way back to the Big Bang. So that's 10 billion light years. There are objects which we think are stuff falling into supermassive black holes, the quasars, which we can see within one or two billion years of the origin of the Big Bang itself. And I showed that funny looking diagram, then I said, oh this is the glow from the Big Bang itself. We do see in all directions a faint glow that's the leftover of the vanished glory of the Big Bang. And that is just a few 100,000 years after-- that was released-- that light was released a few 100,000 years after the origin, after the beginning of the expansion.
So we have a pretty good view up to sort of seven eighths of the way back. Then we have a blank. I didn't mention any objects from there to the microwave background. And it's that era in there where we really don't have any information but that involves the first stars, the first lumps that formed and so on.
That is a kind of observational frontier where we'd like to think of ways to do that. And I think over the next 10 or 20 years, people have a lot of ideas about how to do this, and I think we really will make progress on this era of the very first stars in the formation of the very first galaxies. But that's a piece that's missing from this overall picture it's the early childhood of the universe. Yeah.
AUDIENCE: When you're looking so far into the past or in such great distances, how do you know these observations are making artifacts of dust or who knows, maybe this dark matter somehow interacting with light and causing an apparent change in the light?
PROFESSOR BOB KIRSHNER: So the question is how do you know there is something happening to the light on its way from the distant object to us? Couldn't it be absorbed by dust, for example? And this is one of the things we worried about a lot.
But the interesting thing is that these observations with the Space Telescope that I talked about right at the end, the sign of the effect changes. So instead of the supernovae being a little dimmer the way we see in the region out to about five billion light years, if you go to 10 billion, those supernovae revert-- it reverses, and the supernovae become a little bit brighter. This is what you would expect if the universe has this deceleration and acceleration, but it's not what you would expect if it was accumulated dust along the line of sight.
So we're very concerned about the kind of problem that you've brought up. We don't have tremendously powerful arguments for all of these things. But we certainly are aware of it, and we're certainly trying to make tests. And I think this very distant set of supernovae is something which is quite persuasive that we're not off on the wrong track.
Now each of these different methods seems to converge on more or less the same solution. But we shouldn't be too lazy, and we shouldn't just lean on the other methods because we should do the best job we can with the supernovae themselves. And that's what this space mission that we're talking about is intended to do. Yeah, in the back.
AUDIENCE: On your graphs you focus solely on type 1a supernova [INAUDIBLE].
PROFESSOR BOB KIRSHNER: So the question is I talked about type 1a supernova. I ain't even sure I said that. But anyway why didn't I talk about the other kinds of supernovae, the ones that Hans Bethe had worked on?
Well, the other super-- all the supernovae are interesting, and they all have important contributions to the elements in your body. I want to be clear on that point. But for measuring distances, the type 1a have been extremely good because they have a narrow range in brightness. The brightest ones to the dimmest ones is about a factor of 3. And if you measure the light curve shape-- which I didn't talk about but if you come to my colloquium tomorrow I'll talk about it a lot-- you can figure out to about 10% what the distance is.
So the type 1 have this wonderful property that there are very homogeneous and that the light curve shape helps you make that even better, make them even better standard candles, better for measuring distances. But type 2 are not quite as bright, so you can't see them as far away. They are more varied, so you can't just use the brightness. But we understand them better, and maybe we will be able to use the type 2 supernovae to measure distances at least over some fraction of the cosmic distance scale. And that would be a good thing to do to get an independent measurement of the Hubble constant, the rate at which things are expanding now. So that's the answer. Yes.
AUDIENCE: So we've changed from a decelerating universe to an accelerating universe, so three-part question.
PROFESSOR BOB KIRSHNER: Uh oh.
AUDIENCE: First part, what changed? Second, is something fundamentally different about the physics in a decelerating versus an accelerating universe. And third, is there something [INAUDIBLE] that we could shift again or at least [INAUDIBLE]?
PROFESSOR BOB KIRSHNER: So the question is if we've switched from deceleration to acceleration, what happened at that point? And what does that imply? So the idea is that you have dark energy and dark matter. They are in a kind of tug of war.
And what really matters is the energy density, the density in each of these. If it's really the cosmological constant-- which is not guaranteed-- that is the dark energy, then you can think back into the past-- what was the energy density, the same. For the dark matter though, as you go into the past, if you take any chunk of the universe, it would have been smaller, so the density would have been higher.
So there's nothing magic. It would be the same pair-- the same contest going on back in the decelerating time. It's just that as time goes by, the density of the matter decreases enough so that instead of being able to slow things down now, it's being propelled by the dark energy.
The thing that is a little pecu-- and then the question of what will happen in the future depends on whether the dark energy really is the cosmological constant or some other more interesting thing. But we don't know that for sure.
So we should be a little cautious about the future. We should-- the past is the part where we can actually make some observations. But there is something very peculiar here, which is that this change from slowing down to speeding up took place not very long ago, maybe five billion years ago. It seems like a long time, but it isn't really in this whole cosmic scale.
And it seems like a little bit of a coincidence that somehow that happened about the time we're-- about now that the average-- the value of the slowing down stuff and the speeding up stuff seems to be about the same. That seems peculiar. I made a big deal about us not being at the center, and we should be modest about all this stuff. But it does seem like picking out a special time, the time when the Earth formed and the sun formed as important somehow in the history of the universe. That smells a little funny to people, and it makes people think maybe there's some deeper connection between the dark matter and the dark energy so that they really are always or should be of the same order of magnitude.
So there's a lot we don't know. And that is one of the points that has made many people curious about what the dark energy might be. one more in the back.
AUDIENCE: [INAUDIBLE] lecture, I think that you should [INAUDIBLE] also. So what's the speed that the universe is going, accelerating, and does the universe have a shape like--?
PROFESSOR BOB KIRSHNER: So the question is what's the speed at which the acceleration is taking place. Well, the expansion is, roughly speaking, a part in-- things are getting bigger by a part in 14 billion every year, not physical things in the solar system, but the distances between the galaxies are growing by 1/14 billionth each year. So it's really subtle. You need big distances to be able to get this straightened out.
The Big Bang of course is in all directions, so the shape of the universe is a tricky thing. But the geometry of the universe is the geometry of flat space. This is quite-- this is not the only possibility. Einstein's theory of relativity explains how the presence of matter can curve space, and you could have different kinds of curvature. You could have the curvature like a sphere, the curvature like the saddle.
But the curvature we've got is very nearly no curvature. That is the Euclidean geometry seems to work. The circumference of a circle is 2 pi times the radius. Well, that's true in flat space, and it turns out to be true in the universe that we actually live in.
The evidence-- the best evidence for that comes from this glow from the microwave background and the ripples on that where the angular size of that is something that gives us a very strong measurement of the geometry of the universe, and it seems to have the properties of flat space. So that-- many theorists expected that, but it's very good to have a measurement that shows that that's really true. Well, thank you very much. I really appreciate your attention.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact firstname.lastname@example.org if you have any questions about this request.
Light from exploding stars halfway across the universe reveals an astonishing fact: the expansion of the universe is speeding up! Astronomers attribute this to a mysterious "dark energy" that drives cosmic acceleration. And we need a lot of it—dark energy accounts for 2/3 of the matter and energy in the universe today. Curiously, when Albert Einstein first thought about gravity in the universe, in 1917, he introduced a repulsive "cosmological constant" that he thought would match a static, unchanging universe. When, in 1929, astronomical observations showed the universe was not static, but expanding, he stopped talking about the cosmological constant. It has dubbed his "greatest blunder." But today's observations show that we need something that acts just like the cosmological constant to produce cosmic acceleration.
Robert P. Kirshner, the Harvard College Professor of Astronomy and Clowes Professor of Science at Harvard University and president of the American Astronomical Society from 2003 to 2005, describes how we use observations of supernovi—exploding stars—to trace cosmic history and to learn more about the nature of the dark energy, one of the deepest mysteries of the physical world.
The Bethe Lecture Series, established in 1977 by the Cornell Department of Physics and the College of Arts and Sciences, honors Hans A. Bethe who joined Cornell's faculty in 1936, and whose research extended across fields as diverse as the quantum theory of solids and the nuclear processes that power the sun, receiving the Nobel Prize for the later work in 1967. Bethe continued to make significant scientific contributions until his death in 2005.