Eric Linder

Pavlish:

It is August 1, 2007. I am here at Lawrence Berkeley Laboratory to interview Eric Linder. First, I would like to ask you, Professor Linder: what are the milestones or turning points in the Supernova Cosmology Project.

Linder:

First I should make it clear, that I am not actually a member of the Supernova Cosmology Project. I am a theoretical physicist. I specialize in cosmology, gravitation, and various aspects of the early universe. I work very closely with other people here at Lawrence Berkeley Laboratory, including the Supernova Cosmology Project people. But, I might be a little different than the other people you have interviewed, in that I both have the close ties from knowing the principals of both this team and the High-z Supernova Team. I have known, for example, Saul Perlmutter and Bob Kirshner since the early 1980s, well before these projects got going. Also, being a theorist, I have somewhat of an outside perspective on the development of the supernova method to tell us about the early universe. Actually, when I was a graduate student, I started off by working on supernovae. I worked on the core collapse supernovae, the type II supernovae, trying to use those as distance indicators. Then, I went on to other things. For about ten years, from the mid 1980s to the mid 1990s, I was more on the critical side of saying, “I think it is quite difficult to use type Ia supernovae as distance indicators.” Then, gradually as the data came in, and the analysis methods and the technology developed, I was transformed from a critic to a believer, that yes, they have done the experiment carefully enough, they have done the analysis strongly enough, that this is a legitimate result, the acceleration of the expansion of the universe.

Pavlish:

Was it the supernova results that took you from being a critic of the supernovae as distance indicators to being a believer? Or was it other independent work?

Linder:

There were certainly indications from other lines of reasoning that a cosmological constant could be a constituent of the universe. But certainly the one that I think was most convincing, were the supernova results that came out from the groups in 1998. There has been a long history in cosmology and astrophysics, where people think that the cosmological constant can solve certain problems or explain certain observations very well. You may well know, that in 1975 a paper was published called “An Accelerating Universe,” by James Gunn and Beatrice Tinsley. It is a very interesting paper to read, in that they showed how the observations they made using the properties of galaxies implied that there was a cosmological constant. But they also did a very careful analysis of where this line of reasoning could break down. One of the things they identify, the evolution of galaxy properties, turned out to be indeed very significant. So, they did not discover the accelerating universe. If you took into account the evolution of galaxies, it was not convincing in that way. In the mid 1980s, there was a paper by Edwin Loh and Earl Spillar, saying that the matter density was exactly critical, and that there was no cosmological constant at all. Again, people very quickly looked at the data and looked at the analysis, and found that again evolution was a problem and you could not draw that conclusion. I think, when the supernova results started to come out in the mid 1990s, people again knew what to look for. They said, “Have you tested all the important things to make sure that the data is convincing?” I think, that by the 1998 results, it was convincing to me and it was convincing to a lot of other people. In those things, there is no eureka moment, such that no one was thinking about a cosmological constant and then suddenly someone said that a cosmological constant exists. The thoughts had been there all along and people had even tried to make it work with observations before. But, this was the one that was actually successful in getting good enough data and a careful enough analysis to convince people.

Pavlish:

Going back to the early 1980s, you say that you knew both Saul Perlmutter and Robert Kirshner? That was through your involvement in theoretical physics?

Linder:

It was mostly because I was working on these Type II supernovae, using what is called ‘The Expanding Atmosphere’ method for doing distance determination. Robert Kirshner had worked on that in the 1970s as well. He visited Stanford University where I was a graduate student. We talked about his work and about the work that our group was doing at that time. Of course, Saul was just across the Bay here at Berkeley. We would have meetings back and forth between Stanford and Berkeley on topics of interest in Astrophysics, including the supernovae.

Pavlish:

That was in the early 1980s, when the Automated Supernova Search, the Robotic Search, was just getting started here at LBL?

Linder:

At that point, it was very much work one by one, finding one supernova at a time and trying to learn about it. The sophisticated searches that were developed in the later 1980s, 1990s, were not there yet. People were talking about the possibility of getting the technology to the point where we could do that. But, remember, even CCD detectors were not all that common in the early 1980s. By the later part of the decade, they were getting much more common.

Pavlish:

Where were you around 1998? Is that too much of a skip forward?

Linder:

We can cover that, and then skip back if we want. In 1998 I was a senior researcher at The University of Massachusetts, at Amherst. As I say, I was doing a number of projects in theoretical astrophysics and cosmology. I saw the results that were coming out on the supernovae, and I was extremely interested in them.

Pavlish:

How did you see the results?

Linder:

I am sure that at some point I saw them on the online archive, on the AstroPh archive, and in the journals. It is possible that I might have been hearing about them at conferences. But I do not have a specific memory of a particular conference where I heard about them. I had always been very interested in the ingredients of the universe and in a cosmological constant. For example, I was a Princeton Undergraduate where I did a Junior Paper with James Gunn and James Peebles on the cosmological constant. Then, in the mid 1980s, a part of my thesis work with Robert Wagoner at Stanford was looking at what we now call different equations of state. Things can behave like matter, and they can behave like radiation. But the idea that things can have a negative pressure, which would have to accelerate the universe, I did a lot of what is called phenomenological work. Without having a clear idea of an origin model, saying: if someone gave you something with a negative equation of state, how would that affect cosmological observations? How would that affect the expansion of the universe? I wrote a number of papers in the 1980s, on that. Then, that was one of the things that I put into my cosmology textbook, First Principles of Cosmology, which appeared in 1997. It did actually predate the observational discovery. One of the things I was interested in, in the papers from the observational groups, was in how they were getting the matter density, cosmological constant density, and starting to look at the equation of state, saying that maybe it is not just a static cosmological constant. Maybe there is some dynamical physics in it. Historically, you may know the work done on the dynamical scalar fields, the Peebles-Ratra work. Have you encountered that? The idea was that the cosmological constant is unchanging in time. That raises certain issues, such as why should it be important just today? Why was it not important early on? Has it been always the same? When I was at Stanford, one of my classmates was Bharat Ratra, who went on to work with James Peebles at Princeton. Talking about a dynamical origin. Instead of a cosmological constant, you have what is called a rolling field. It does evolve with time. That does solve some of the problems. I had been interested in looking at that work. All of this had been in mind in that decade, from the mid 1980s to the mid 1990s. So, once observations actually come out, showing that this might be what our universe is doing, I got very interested in working in this area again.

Pavlish:

The results indicate too that there might be a changing Lambda?

Linder:

The results did not say that, but it raised the question again, that: given that there is something with negative pressure, something that accelerates the universe, is it just an unchanged Lambda, or we do have that alternative of saying that maybe it is a varying field. Maybe it is with connections with the early universe of High Energy Physics. It opened up all of that area that I had looked at ten years before. I said, okay, people will now listen to me when I talk about it because there is some observational evidence as well.

Pavlish:

I have two questions in my mind as a result of what you have been explaining to me. The first is, I am interested in the textbook aspect of this. Did you have to revise your textbook when the results came in?

Linder:

Actually, I wrote most of the textbook in 1991. I moved to England and then the textbook got published in England. It was not until 1997 that the textbook came out. It was not revised as a result of the observations. I did not have the chance to revise it. Now, I am thinking that maybe I should update the textbook at some point. The observations today are coming rapidly enough that I want to wait for things to settle down, to have a better idea before putting it into a textbook again.

Pavlish:

Your textbook is more of a theory book?

Linder:

That is right.

Pavlish:

You would not say in it that the universe is accelerating? I have seen some college textbooks that talk about the deceleration of the universe.

Linder:

That is right. There is a section on the fate of the universe, in which I talk about how for universes with equations of state, which are more positive than minus 1/3rd, which is what people considered throughout most of the 1900s, that you do have a simple connection between geometry and the fate of the universe. But then, I also talk about equations of state with Sigma, which today is called W, causes faster expansion. I have pictures of what it would show if you went from a deceleration to an acceleration, which is similar to the sort of picture that I am sure that you have seen a number of times today. There is certainly the idea that when you have these equations of state, the situation is complicated because the geometry does not give you the destiny. That, there is a more complicated relation. Yes, this book did talk about these equations of state, but we did not have any observational evidence for them at all. I talk a little bit about some of the supernova magnitude tests, what that would look like for different cosmological models as you put in different equations of state. This was an old test from the mid 1980s, for number counts which was shown to have a problem with evolution of objects. For the supernovae, I did not have data points at that stage to put in the textbook at all. I was very pleased when these observational groups were able to show that this could be a probe of the cosmology.

Pavlish:

Your textbook is on a different plane from some of the ones I have looked at, because it did not require revision after 1998.

Linder:

It certainly required an update as far as the observational situation. But the concepts were already there.

Pavlish:

You do not actually teach the students, for example, that the universe is decelerating in your textbook.

Linder:

That is right. It is an open question in the textbook.

Pavlish:

The other question I have: you worked with James Peebles. I was thinking about the Great Debate of 1998 between Turner and Peebles, in the fall of 1998, on the question: Is cosmology solved? I think that the supernova measurements had some impact on the way that debate went. I was wondering if you were there, did you follow it?

Linder:

I was not there. It was a very exciting thing, to have the supernova results, and early hints of the Cosmic Microwave Background result within a year after that. Again, there were some researchers on the Cosmic Microwave Background at U. Mass Amherst, Phil Mauskopf who was working on the Boomerang experiment. We heard some early rumors that they were finding that the geometry of space is flat. If you accept, as I think a lot of people did at that point, that the matter density was not sufficient, that it was not an Omega Matter of one but that it was less than one, if you add to that the flatness, then that is an independent line of evidence that there has to be another component. It does not tell you that it has to be an accelerating component, but it tells you that there has to be something else that we do not know. In that sense, I had early rumors that in addition to standing on its own, it was getting support from other areas of astrophysics as well.

Pavlish:

That was around the summer of 1998?

Linder:

I believe that it was around the summer of 1999.

Pavlish:

Can we put a stamp on when the discovery of the accelerating universe happened, in your view?

Linder:

[laughs] Again, it is difficult. Theorists were talking about it since Einstein. It depends. What do we mean by discovery of it? That our universe is indeed accelerating? I was pretty much convinced at the time when the papers came out in 1998 [and 1999]. There were certainly some open questions, but it was clear how you would go about answering those questions. Again, I was very excited that the universe added a level. That there was more to learn about the universe, and that the things that were more to learn about the universe, might have connections to the very highest High Energy Physics, might have connections to the nature of gravity, certainly would affect the fate of the universe; that these were all very interesting, very big questions to work on. I certainly cannot give a month and a year when I considered that the accelerating universe was established. As I say, I think that the paper that came out were very influential. In particular, I thought that the analysis in the Perlmutter et al. paper addressed certain issues that had made me most skeptical. One of the things was this idea of a clumpy universe. That, in the Friedman-Robertson-Walker model, everything was smooth and the distances that you calculate in models are smooth universe distances. If matter is clumped together, for example if Dark Matter were in clumps, then a light ray going through the universe does not really feel the entire smooth density. It is avoiding the clumps. If it hits a clump, it gets absorbed. So it is avoiding the clumps, and so it is not seeing the same distances as the smooth distances. That had worried me in the ten years that I was critical of the supernova work. In the Perlmutter et al. article, they actually went through this analysis and showed how much it would affect their results. Even though it did affect the result, it did not affect the result so much that it would remove the accelerating universe effect. As I say, there were a number of things, and in particular theoretical analysis. Not being an observer, I cannot really comment on the data quality. But once you have the data, how rigorous an analysis you do can vary. There were things that I liked in particular in that paper. Again, having both papers coming to consistent conclusions was a major factor in convincing me and I imagine a lot of the community.

Pavlish:

I have a question about how experiments end, that relates to theory although it is a question more directed at experimentalists. If you would give the theorists’ view on it, that would be nice. In his book, How Experiments End, historian of physics Peter Galison tries to establish something more sophisticated on how experiments end. Thomas Kuhn and his group would say something like, experiments end according to a Gestalt picture model, when they fit into a preconceived theoretical schema. The social constructivists would say that experiments end when they mesh with the interests of the dominant theorists. For Galison, neither of these accounts seems satisfactory. In your opinion, when do experiments or observations come to their conclusions? How does consensus form? I might suggest that an experiment ends when everyone in a research group agrees with the result, or when it gains general acceptance. I am rather naïve about the role of theory.

Linder:

Here it depends exactly what you mean by an experiment. Certainly, a given project to collect certain data at a telescope and so forth, that ends one way. But if by experiment, you mean: when do we decide that the cosmological constant exists, or something like that, that is a very different answer. Since I am theorist, I will address the second version. My answer would be, that it never ends. Take Einstein’s Equivalence Principle. Today they are still testing Einstein’s Equivalence Principle. They tested it to one part in ten to the eleventh. Then, to one part in ten to the twelfth. There are proposals now to test it to one part in ten to the fifteenth. Even further, future ones, to one part in ten to the eighteenth. As long as there is some moderately convincing theoretical reason why there might be some breakdown in symmetry, violation in some kind of conservation, then people will want to continue testing it. There is of course the issue of whether the technology is ready to test it, and is it affordable to do so. Even with things like the value of Newton’s constant, where you are not really testing a theory, you just want to measure the value, people are still getting better and better measurements. Something like: is the cosmological constant the solution? I think that is going to just keep going on and on and on. There is this amusing story, about the physicist Wolfgang Pauli, who was very instrumental in quantum field theory. You may well know this story. He dies. He goes to heaven. He meets with God. God says: you have been investigating nature for so long. You have done such a great job. Is there any question, any mystery that you still want to know the answer to. I will give you the answer. Pauli says: why is the value of the fine structure constant what it is, 1/137? God starts to explain. God starts writing all these equations on the blackboard. Finally, he comes to an end. Pauli is just looks at him and says: still wrong. He just never accepted it. To a lot of physicists, there may not be an answer to why there is the cosmological constant or why the cosmological constant is the particular tiny value that we measure. At least for theorists, they never want that experiment. There are always practical concerns. My answer would be that there is always more to learn. We asked the question, what is the value of Omega Matter, the matter density? We got an answer and we said, okay, now that means that there is also a cosmological constant. That leads to the question of: Is it really a cosmological constant? Is it static? Does it change with time? Is there an equation of state different from minus one? Then, we started asking: Is it a constant equation of state, or does that vary, itself? We are just going to keep asking these questions. The theorists do not want to put an end to it.

Pavlish:

Did you start considering the cosmological constant to be Dark Energy at a certain point, as a more general term for it?

Linder:

Dark Energy, the name, was coined by Michael Turner. In looking at the properties, in the 1980s, asked what does a negative equation of state do? Go back to the question again, please.

Pavlish:

When did you start thinking of Lambda as Dark Energy? Or, do you? Maybe you still do not think of it as that.

Linder:

It is a possible explanation for Dark Energy. I am certainly not convinced that the answer to Dark Energy is Lambda. There are two issues. One is, if we figure out what Dark Energy is, and it is not Lambda, we still have not solved the question of what happened to Lambda. Why is it exactly zero? That is, in some sense, a separate question from Dark Energy itself.

Pavlish:

By Lambda, you mean the quantum vacuum energy?

Linder:

That is right. Why is the ground state exactly zero? If it is, or even if it is not exactly zero, then what sets its value?

Pavlish:

Einstein himself, when he put it in his equations, was not necessarily thinking of it as that.

Linder:

That is right. You can think of it from a gravitational context, as well. There you still have the question of, if you can put it in the Einstein equation, what prevents you from putting it in. As I say, it really is a separate question from Dark Energy. We will have to answer both questions. First, what is causing the acceleration of the universe? If it is not Lambda, then we also have to answer: Well, what about Lambda? I think of them as two separate things.

Pavlish:

I must be misunderstanding something. My impression was that the acceleration of the universe was discovered by these experimentalists. The simplest answer would be that it is because of the cosmological constant. But, it could be due to a number of reasons. It could be because of quintessence or the quantum vacuum fluctuation. My impression was that Dark Energy would be a general term for what we do not understand. It could be any one of those things. It is a more general term.

Linder:

That is right.

Pavlish:

But what you are saying, is that even if we find out what Dark Energy is, we still have to figure out what Lambda is.

Linder:

That is right. If Dark Energy is a quintessence field, then we have to explain why isn’t there a Lambda in Einstein’s equation? Why isn’t there a Lambda arising from ground state vacuum fluctuations? That is a separate issue.

Pavlish:

You are saying that having an accelerating universe is not the same as having a Lambda in Einstein’s equation?

Linder:

Lambda would be one possible solution to Dark Energy.

Pavlish:

Right.

Linder:

But, if Dark Energy is something different than Lambda, we still have to ask: what did you do about Lambda? We have to either set it to zero by some symmetry principle, or we could have both. We could have a Lambda and we could have a quintessence field.

Pavlish:

That is helpful.

Linder:

It is complicated.

Pavlish:

Is there anything else that you would like to say for the record?

Linder:

All I should say is that I was very excited when the discovery was made. It was shortly after that, that I wrote to Saul, and said, “I think this is very interesting. I understand that you are planning a new experiment at the Space Satellite to investigate this. I would like to add my skills to that.” That is how I ended up here at Berkeley Lab. As far as the actual discovery, in that period of the observational groups, the Supernova Cosmology Project and the High-z Supernova Search Team, I did not have any direct contact with them at all. Now I do. The next Supernova Cosmology Project paper that is going through its draft form now and is going to be published soon, I am a part of that. But I am not directly a part of the SCP.

Pavlish:

You were so convinced, that you actually wanted to join the team!

Linder:

That is right.

Pavlish:

Thank you very much. Thank you for your time.