Oral History Transcript — Dr. Harvey Tananbaum
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
See the catalog record for this interview and search for other interviews in our collection
Interview with Dr. Harvey Tananbaum
Harvey Tananbaum; June 26, 2002
ABSTRACT: A biographical interview; Tananbaum was director of the Smithsonian Astrophysical Observatory's Chandra X-Ray Observatory at the time of interview. Discusses his childhood and education including time at Yale and MIT; initial forays into X-ray astronomy; anecdotes about Riccardo Giacconi and launch of Uhuru Satellite in 1970; discovery of first black hole in Cygnus S-1 and confirmation of binary accretion model as source for x-rays. Transition of AS&E staff to Center for Astrophysics; building of High Energy Astrophysics division at CFA; Tananbaum's research interests; CHAMP project with Chandra. Discussion of building Chandra as well as comparison of Chandra project with Hubble telescope. The politics and finding for Chandra; its operation. HEAO-B (Einstein) mission covered including its precursors, detector technologies, and science contributions. Tananbaum's time on various NASA committees and the creation of a long-term strategy for space astronomy mentioning Chandra, Hubble, SIRTF, and NGST. Discussion of important topics in astronomy research including the merging of physics and astronomy in some areas. Committee contributions including decadal survey work. Thoughts on religion, family, values and personal meaning. Most significant changes in his career including personal rewards.
Session I | Session II
McCray:Today, Actually, I have two topics. I wanted to talk about the Einstein mission because we missed it last time. And then I have some more general questions about X-ray astronomy, cosmology, and your views about some of the current relations between astronomy and physics. We talked last time about how you left AS&E and came here, but we didnít really talk a whole lot about how HEAO-B came about as a mission. I know it has been covered a bit in some other sources, but I just want to get your perspective on how it came about.
HEAO-B. I think, again, the Tuckersí wrote a book The Star Splitters which covers the High Energy Observatories. So one can always use that as a reference to check a date or two against what could be, at this point, a little bit of faulty memory. The genesis of the program really goes back to 1963 in the sense that a year after Sco X-1 and the background were first observed, Giacconi wrote a proposal or a White Paper. It talked about an X-ray satellite and it talked about an X-ray telescope. In the late 1950s, he and Bruno Rossi had looked at possible designs for an X-ray telescope. They found in the literature a work that Wolter had done. He was a German scientist and he was interested, I think, more in an X-ray microscope. I would say it was back in the 1950s. The problem was that the microscope would be very small and the properties of the optics that you would need—the slope of the glass or whatever you made the microscope out of, the smoothness, the local roundness, the local slope—on a very small object would be tremendously demanding.
And so, it didnít make a lot of sense as an X-ray microscope. But those same designs of cylindrical optics, parabolic and hyperbolic surfaces where you could use grazing incidents to focus X-rays, when you scaled it up to a telescope, it became quite attractive as a design. So Bruno and Riccardo published a paper about how you could make an X-ray telescope in 1960. In 1963, Riccardo wrote this White Paper. There may have been co-authors, but it was primarily Riccardoís leadership and his ideas and vision of the future. Talked about an X-ray telescope about 1.2 meters in diameter with arc-second angular resolution and a long life as a logical step to be headed towards to do meaningful science, meaningful astronomy in the X-ray band. NASA had a lot of trouble with this proposal. I mean, there was one rocket discovery flight, a flight or two more flights, one source, second source, a little background. And here is somebody proposing a program for the next few decades.
So they did start some studies of what eventually became Uhuru, a satellite to make a map of the whole sky with some reasonable amount of sensitivity. And they also funded some technology work and I think, I donít who precisely, but I think that there was a sense that perhaps it might make more sense to build an X-ray telescope and to look at the sun first because the sun is much closer. It is a relatively modest X-ray source as stars and other objects out there go. It is 1028 ergs/sec in X-rays. Sometimes 1029, sometimes 1027 depending if itís flaring or quiet. Sco X-1 is like 1037 or 1038. So it is 109 times or a billion times more luminous as an X-ray source. Of course, it is a different physical process. The accretion onto a compact object is much more efficient than the twisting and breaking of the magnetic fields in the corona of a star, which is how the sun generates its axis. In any case the point was the telescope was going to be a real adventure and the suggestion was to develop one to look at the sun first. So AS&E started working on that technology. Pippo Variana joined the group from the University of Palermo.
McCray:What was the personís name?
Pippo which is a diminutive for Giuseppe, I guess. He passed away about ten years ago, maybe a little bit more now. A relatively young man with a heart attack, heart disease. But in any case, he came to work at AS&E probably in the mid-1960s There was this technology effort underway to build a telescope which was actually flown on a rocket and took some beautiful pictures of the sun followed by an actual telescope on the Apollo, the Skylab. It was called S-O54, I think that was the telescope that AS&E built and got these marvelous pictures of the sun and were able to look at flares. And really determined that the magnetic field and the dynamo kind of a process, the rotating of the magnetic field, the breaking of the magnetic field lines was the underlining physical process that accelerated the particles and produced the X-rays. So it was really a tremendous project. Well, in the late 1960s as this project was going forward to study the sun enough progress was being made that a proposal, I think possibly an unsolicited proposal, was sent in by a group led by Riccardo.
It was called the Large Orbiting X-ray Telescope (LOXT). Leon van Speybroeck got involved in studying the optics at that point and really beginning to work with the engineers on the design of what would be this 1.2 meter, 6 nested surfaces, X-ray telescope to study stars, supernova remnants, binaries, the extra galactic objects. This was before Uhuru had even flown, so they didnít know about the extended emission in X-ray clusters. We didnít know about the underlying mechanism of accretion onto a compact source in a binary system as the fundamental process powering the bright galactic sources. But anyway, the technology work got started and there certainly was a call for proposals, the HEAO program was structured such that this telescope was going to be a sort of a centerpiece of a series of four missions. One of which would be a more modest telescope, another of which would do gamma rays rather than X-rays. The cost of the program started escalating in the early 1970s. There was some cost growth from the Viking program to study Mars with landers. The NASA budget was squeezed. Weíre talking about the time when the Vietnam War was heating up.
McCray:Apollo was winding down.
Tananbaum:Yes, Apollo was winding down and NASA decided in January 1973 shortly before we got here to cancel the HEAO program. Riccardo and few of the other senior members, probably Herb Friedman and others, were working with NASA to reconstruct the program as a somewhat smaller, more modest set of missions so that it could be afforded and go forward. And so, later in 1973, it became three HEAO missions with fewer and smaller instruments. The LOXT was scaled down from a 1.2 meter to 6/10th of a meter telescope. HEAO was to be based on more of the already existing spacecraft to keep the costs more modest. We were using propellant to unload momentum and the lifetime of the mission would be a year or two at most. So it met only some of the visions that Riccardo had, but as it turns out, it was a fantastic step forward for X-ray astronomy. So it met some of the visions that Riccardo had, but as it turns out, it was a fantastic step forward for X-ray astronomy.
McCray:Because it could image?
Tananbaum:Because it could image. But it took us AXAF/Chandra to actually accomplish the vision that he laid out in 1963 and what the LOXT was originally intended to be. So there was this series of proposals, and technology work, and then reconstruction of the HEAO program. At this point around 1974, the HEAO-B, the second of the three HEAO missions is what the X-ray telescope became.
McCray:How did the results from Uhuru affect the plans for HEAO-B as people began to understand binaries?
Tananbaum:I think the results from Uhuru, of course, strengthened the case for having an X-ray telescope because with Uhuru we were seeing the brightest few active galaxies. We were seeing the brightest dozen Seyfert galaxies. We were seeing extended emissions from clusters and we clearly wanted more detailed pictures. We had, I think, made the case that the science was exciting. Neutron stars and black holes and binaries, for example, at least meant that there was some understanding of the physical properties that were generating the X-ray emission. So I think in one way it created a broader interest and a more enthusiastic science case. It certainly affected the things that we looked at with Einstein once it was built and launched. We knew which clusters were the brightest clusters from the Uhuru data. We knew which ones would be the most interesting to map. We could actually estimate what exposure times would be needed to get a meaningful image or spectrum. I think the technology was almost totally different. There was no telescope on Uhuru. There were no imaging detectors.
McCray:How are imaging detectors different?
On Uhuru we had proportional counters which, if left to their own, would look at the whole sky facing them. We had mechanical collimators which were slats that looked sort of like a long series of tubes side-by-side and the walls were absorbing the X-rays. So if you looked within, in one case, a Ĺ degree of a source and in another case 5 degrees of the source, the X-rays could come down the tubes. If a source was off to the side, the X-rays would hit the walls of the tubes and be absorbed. And as you actually swung in a circle, you would go past a source and you would see a signal build up in a triangular shape to a peak and then come back down with the triangle determined by the opening of the tube relative to the length of the tube. So it was just mechanical collimating devices that allow you to determine from the peak in the signal the direction in which you are looking when you saw a source. If you cross the source from a few different directions, you could take the peak of the first triangle and draw a line with an uncertainty in the sky. And in another direction draw another line and use the intersection of the two lines tells you where the source was. And if instead of getting a sharp triangle the thing was flat in the middle or had a tail to it, you could tell the source had a finite size that was measured in fractions of a degree.
You couldnít see things on an arc-second scale because the device wasnít nearly that precise. So the detectors themselves didnít have an imaging property at all, but these mechanical collimators allowed you to determine the source directions and some estimate of extent. Now with a telescope, you actually will cover a 1/4 to Ĺ square degree kind of a field, youíre actually imaging just like you would with an optical or a radio telescope. Although with grazing incidence, X-ray optics, the technologies and the tolerances are different. Now you want to bring it down to a detector which allows you, in the case of early rocket flights, to look at the sun. They flew film; very X-ray sensitive film. They would step the film in frames every second or so and then you would look at the image of the sun one second, a second later, a second later. Well with a satellite, the film would be pretty hard to recover, although I guess the military have found certain ways to do that in some reconnaissance missions. So there were two kinds of detectors which we developed for Einstein.
One was based on gas proportional counters, which were the basic technology that we flew on Uhuru, but we were adapting from the physics community. We were going to have multiple wires in the counter and so the location of the event would be determined by what wire or set of wires got a signal; it was a 2-dimensional imaging device. And so from the wires that gave you the signal or the signal over a couple of wires that were centroided, you could determine where in the counter the event occurred. Now an issue with multiple wire proportional counters was challenging the technology in the 1970s with these wires so close together. You had high voltage on them meaning you sometimes had arcing or a breakdown of the counter. And if the counter rate went very high, the arcing could get even worse. So it was right on the edge of what one could do, but they were called imaging proportional counters and they were multi-wire counters.
McCray:Where were they built?
Tananbaum:We built them in-house. Paul Gorenstein and Rick Harnden were the lead scientists working on those. The other kind of detectors we also built in-house, they used a micro channel plate which are very, very small diameter glass tubelets, many, many of them packed together so that it looks like, probably doesnít look all that different from a bundle of fiber optics. The diameter of the glass tubes was 10 to 20 microns. There were millions of them packed together in a bundle to cover a 1/2? field of view.
McCray:These were made in-house?
Tananbaum:We bought the micro channel plates from a company called Galileo. Itís in Sturbridge, Massachusetts. It had a different corporate name when we bought the channel plates from them in 1970s. It almost certainly did. Galileo has been bought by Thompson and so maybe it was Galileo then. The trick wasnít so much that, the channel plates were challenging, since it was technology that had been developed also for the reconnaissance satellites and night vision—
McCray:Yes. Thatís a question that I have.
Tananbaum:So there was some crossover in the technology in the beginning when we worked with the devices. We needed special permissions, letís say, to have access to that kind of information. But by the time we flew them, they were considered in the open area.
McCray:Who would you go to get the permissions from?
Tananbaum:We would go through a standard clearance procedure. I was being a little facetious with the permissions. We had to have clearance, security clearances. And because AS&E in the early days had been doing weapons testing, most of the people who worked there had such clearances anyway.
McCray:So that didnít present any complications?
Tananbaum:No. It meant that there were certain things on the micro channel plates that werenít in the literature. If you didnít know what was going on at the night vision labs, for example, you might not be aware of the most recent work on channel plates because it wasnít always being published.
McCray:Iím confused about that because when I think of night vision, I think of infrared.
Tananbaum:Yes, but these channel plate detectors could be used as infrared detectors too. What we did differently is we put a coating on the front surface of them to enhance their ability to detect, which means absorb an X-ray and generate an electronic signal. Now what was difficult for us and where the challenge was for Einstein was how to convert the information. An X-ray hit the fluorescent coating which you put on the front to generate some electrons. You had a 3,000-volt electrical field down this pair of channel plates so the electrons got bounced into the walls and kept producing more electrons which got accelerated which produced more electrons. If you have this cascade of electrons coming out the back end—
McCray:Like a photomultiplier?
Yes, except the charge was somewhat localized. Now, how were you going to readout the location? And if you had, letís just say for example, you had a goal of reading a position to 1-arc second. If you have a field of view which is, letís say, 15-arc minutes on the side. That is 60 times 15, which is just under 1000. You may have 1,000 elements, we call them pixels. If it were a true imaging device then also times. Also 1000 in the other direction. So you have 1,000,000 detection elements or pixels. Well, you couldnít imagine putting 1,000,000 amplifiers or a 1,000,000 readouts in there. We ended up back in a scheme which had at first been worked a little bit at the University of Leicester in a rocket program several years earlier using a set of discrete wires that ran in one direction, and then a set of discrete wires underneath that ran in the perpendicular direction.
And you actually let the charge from a single event get into an electron cloud which was bigger than the spacing of the wires. And then you used the signal on, say, three or four different wires to look at the amount of charge on each of the wires and compute a centroid, which actually gave you the position in that direction to higher precision than the spacing of the wires. And we actually were able to demonstrate that we could regenerate the position of each individual X-ray event to the order of an arc second. The telescope itself had a core which was about an arc second, but 50% percent of the photons, the X-rays were probably in a region that was probably closer to 5-arc seconds in size. Similarly with the ROSAT satellite which flew after Einstein. Both contained significant amount of a signal on a scale that was about 5-arc seconds. The advantages of Chandra over those two missions was that the same equivalent signals were contained in about Ĺ-arc second. So in the one-dimensional sense, Chandra has about ten times the concentration of signal which reduces the background on the detector because you were looking at a smaller region. But also allows you to resolve some features that are separated roughly on that scale. From a two-dimensional imaging point of view, we can either make the case that the improvement in capability has really been, if itís a 10 in a linear dimension for two-dimensional, then imaging it approaches 100. I think a fair argument is to say that Einstein and ROSAT were of order 5-arc second. Although at times we talk about doing 1-arc second. We could get source positions in the case of Einstein to 2-arc seconds precision.
McCray:You were the Scientific Program Manager which sounds like a typical nebulous NASA title.
Tananbaum:Itís one we invented ourselves actually, but it is exactly as you characterize it. It was an invention. When they put the team together in either 1969 or 1970, Riccardo formed a consortium. I think he felt that this thing would be too big for just one institution to have full responsibility, a science institution to have a responsibility for all of it. And to build a consensus behind it was important to involve some of the other groups.
McCray:Make it harder to cancel, in other words?
Yes. And easier to support. So Riccardo was the principal investigator. He was the lead scientist who put the whole thing together. He had four people that were called principal scientists. They were Bob Novick at Columbia who at various times went in and out, but it was possible that they would provide a polarimeter. Never did fly a polarimeter on Einstein, but looking for polarization was their main interest. George Clark at MIT, his interest was to do a Bragg crystal spectrometer which did fly. Elihu Boult at Goddard, he was interested in solid-state detector. It was a silicon detector and in some sense it didnít have imaging properties. The telescope concentrated the signal on the detector and you could get better energy resolution because a silicon detector is better than a channel plates or proportional counters. It was the forerunner of CCDs without imaging. And then Herb Gursky who was at AS&E and then came over from Smithsonian. Herb had some of the lead responsibility for the imaging proportional counter, but it was more that the AS&E group, which then became the Smithsonian group, was responsible for the imaging detectors on the telescope.
Riccardo was the PI and then Paul Gorenstein was involved as the lead on the imaging proportional counter as time passed. Leon van Speybroeck was the lead for the telescope and interested in the technology. I mean, he is really the worldís lead on X-ray optics and all the details on making telescopes. Fascinating track record of just success after success. (Note: Leon VanSpegbroock died December 25, 2002 from metastasized melanoma). Then there was the channel plate detector, which was called the High Resolution Imager, HRI, Ed Kellogg, Pat Henry, and Steve Murray were all involved. And I was working Uhuru really 1970, í71, í72. Still had lead responsibilities for running the observing program and doing science with it. We were making the catalogues. Steve Murray, Ethan Schreier, myself had written a lot of the software, analyzed the data from Uhuru with help from a couple of professional programmers. But we had the responsibility for designing the software and getting the data processed. We were still at AS&E so it had to be still 1972-ish. Herb Gursky came by and I guess nominally he was my boss, but at this point I was only working on this directly for Riccardo all the time. And Herb asked if I would be interested in getting involved in the first HEAO mission. Herb and Hale Bradt at MIT had a modulation collimator which is a special kind of a mechanically collimated device that has grids of wires that are offset in a precision kind of way. Sort of works on a vernier kind of effect. It can give you positions much more precisely than a standard mechanical collimator.
And they were going to fly such an experiment on the HEAO-A mission. Herb was interested on having me come on to help build the instrument. From my perspective, it had a lot of similarities to what we had already done on Uhuru and I was really thinking at this point that the future was in the telescopes where we would get this great increase in sensitivity. I didnít know anything about telescopes. So I asked Herb for a raise, but then I said I was going to go work with Riccardo on the telescope. Youíll find that theme a few times in my career that I ask for a raise. I got the raise by the way. Herb was very good about that. Started talking to Riccardo about the telescope. We were still at AS&E and he sent me to Leon to learn more about how telescopes worked. The program manager at AS&E, his name was Arthur Vallas and he used to have a little trouble with Leon because Leon is a bit of a perfectionist. And sometimes the people at NASA would need a report, or there was a meeting coming up, and we would need a progress report. Leon wasnít quite ready to write a report because he was still working on something or calculating something. They actually found me to be somewhat useful in that Leon and I had a very good relationship because I was a scientist, I was trusted just a little bit more than the engineers or the managers.
I could ask Leon some questions that he found interesting enough to answer and usually that would end up getting us to the point where we could get a report together for the manager. So I was functioning as a scientist/manager in some of what I was doing. There was always the threat that I could use with Leon as time proceeded, ďWell, if you donít have time to get to the report, why donít I just hop out and Iíll write it?Ē That was enough to set off a panic because, where the engineers couldnít make that threat to write it, I actually was almost credible, but certainly wouldnít get it right and certainly wouldnít do it the way he would want it to be done. So that was almost always sufficient to get Leon to write the reports that the managers needed. So they didnít quite know what the secret or the magic of the process was, but it was really quite simple. And it was done in a very innocent and honest kind of way. So when we were working on a project for about a year, probably by the time we came over here in the Summer of 1973, it was clear that Riccardo envisioned my being in a position to help him run the whole thing. And since all the real titles had already been given out, we needed to have a little bit of a title so that I could sit in the steering group meetings where the principal investigators or the principal scientists and Riccardo would meet to discuss strategy.
The other guys were called the principal scientists, not principal investigators. The only real principal investigator was Riccardo. So we invented this title, Scientific Program Manager. I guess we had job descriptions, but my job description working for the government is astrophysicist. I donít think ďmanagerĒ ever appeared in there. That was fine. I mean, it gave me flexibility as later work on Chandra did. As long as I was paid reasonably well, I didnít care what the title was as long as I was in a position to do things which were interesting and could help the project. And with Riccardo in the beginning and then later after Riccardo moved on, I was in a position to do the things that I wanted to do. I had, letís say, the support of the people that were working on the project. They were confident in my leadership and that I was going to be there to help. So we had this title and you hit it. You ask a short question, you always get a long answer. I guess youíve learned that over these hours.
But that is what the title sort of meant and thatís what my role was. It was helping where help was needed. Sometimes I was in the lab trying to figure out how to make the channel plate detectors work or keep the imaging proportional counters from arcing. I would go to meetings with Riccardo where we argued at one point that we should augment the momentum unloading system which was based on a gas. We knew the gas would run out in one to two years, but we would augment that with magnetic torquers. NASA had established a budget of, I think it was, $400,000 dollars for each of the three HEAO missions to increase the science. It was sort of an incentive. The science group could have $400,000 dollars to do whatever they could. This is on projects that were really $100 million dollars in cost. It was a small amount of money. When we decided that the thing we wanted to do was add magnetic torquers which would allow you to use the earthís magnetic field to unload momentum from the reaction wheels, because HEAO was in low earth orbit. And you energize the torquers in parts of the orbit where the magnetic field would be oriented in a way just so that the input forces would tend to cancel the momentum build up.
And this is a standard thing used on other satellites. TRW offered to do it. It wasnít part of the basic spacecraft that they were building. They offered to do it on a best-efforts basis for $400,000. So it was a wonderful bargain because, in principle, it would have doubled or tripled the duration of the mission. And ultimately, NASA management, first at Marshall and then backed by Headquarters, said that the costs of the telescope were overrunning, which they were and the $400,000 dollars had to be used to help finish the telescope. In our opinion, and I think correctly so, it was breaking the spirit of that fund because things could overrun anyplace on a project and $400,000 dollars was a modest amount in any case. We had to finish the telescope and get it to work at some level. Anyway, the money was taken away and the mission life became limited by the gas propellant. When it came to operating the observatory, we came up with the idea of momentum management in which weíd pair targets knowing that for part of the orbit the momentum was going to build in a certain direction. Weíd pick a target so the satellite was tipped relative to the perturbing forces in one way and then in another part of the orbit, it would be tipped the other way with another target and to cancel the momentum buildup. That allowed us to more than double the lifetime for the propellant. HEAO-2 was not named Einstein until after launch.
Tananbaum:I think it was nearly the hundredth anniversary of Einsteinís birth March 14, 1879, when we launched in late 1978 and we were studying things where gravitation could be important. He was a famous scientists, so it was in honor of him. He didnít probably care at all; he wasnít alive, of course, anymore. Ask me later, by the way, a little bit about Chandra, the naming, if youíd like. (Note: Chandra is named after the late Indian-American theorist Subrahmanyan Chandra, who first calculated that there is an upper limit to the mass of white dwarf and that stars move massive than a certain amount could not end their lives as white dwarfs. The satellite name was selected via an open contact for which more than 6000 entries were submitted.) But in any case, we wanted to come up with a name. We went through a whole series of things. We suggested, at one point, Pequod as a New England whaling ship or something. And NASA said, ďWell, didnít they know that the skipper was mad?Ē And Riccardo sort of laughed.
McCray:The boat from Moby Dick.
Tananbaum:Yes, exactly. But in any case, Einstein was a compromise. It certainly was a nice choice. In April of 1976, we sent a proposal to NASA to start to study a 1.2-meter telescope that would stay up in orbit indefinitely. At that time we envisioned that it would need servicing the way Hubble was done. But the point is, is that NASA was actually willing to allow funding or provide funding and allow a study to start on what became AXAF and then Chandra even before Einstein launched. Because I think they knew that their unwillingness to put the torquers on and extend the life was a real blow to the science or the lifeline of the science for X-ray astronomy. And so, maybe it was to quiet us down a bit. Maybe itís because they felt badly about the decision they made. Maybe it was a compromise of sorts, but we did turn the setback of not getting torquers into the opportunity to start a little faster on AXAF.
McCray:The Einstein Mission was launched in 1978 and had a lifetime of about two years. In that time, what do you think the biggest scientific contributions were from the mission?
I probably look at the contributions as sort of twofold. Weíll say it each time we talk about a mission. Iíve said it about Uhuru. Iíd say it certainly even more so about Chandra. But one of the key things that the telescope did was actually imaging and having the sensitivity to look at more than just the brightest objects in our own galaxy plus a few nearby extra galactic objects. It helped extend the reach of X-ray astronomy to the kinds of things that optical and radio astronomers were doing, to broaden the community of people, the astronomy community that were interested in X-ray astronomy. It was a very significant step towards X-ray astronomy becoming an integrated part of all of astronomy as opposed to just some interesting thing off to the side. If all we were studying was the accreting binaries, it would be interesting, but it would be just a narrow area of astronomy. But by seeing the coronae of stars, by imaging the X-ray details of supernova remnants, looking at the gas in the clusters, by detecting thousands of AGNs, of quasars and Seyferts as X-ray sources, by beginning to resolve the X-ray background.
These were things that then fit into areas of interest for almost any and every astronomer. Certainly there was this major cultural impact on at least the astronomy community. And then there are the specific science areas that were opened up. The realization that the G-stars, the stars similar to our own sun, were not the most luminous of the X-ray sources. That the acoustical processes probably werenít, that the dynamo was what was important. We saw lots of X-rays from K and M stars for example. We saw X-rays from O-type stars, but a very different process involved in winds and colliding atmospheres. We were able to look at the evolution of the luminosity of quasars in X-rays, and compare it to the optical, and find that there were some subtle differences, for example. The X-ray backgrounds being comprised, primarily, of the individual active galactic nuclei. The study of the properties of the hot gas in clusters. It was recognized that you could use the hot gas as a tracer. It was recognized towards the end of Einstein, but really exploited more with, ROSAT, and now with Chandra because you want the temperature and the gas density profiles.
The gas density profile Einstein could do that but a temperature profile was difficult. The hot gas just wants to expand away and the fact that itís held on to by the cluster gives you measure of the overall gravitational force. And it turns out, of course, that most of that gravitational force isnít from the galaxies with their stars and the like, and itís not from the hot gas because you can measure how much gas and how much starlight is present. Itís from what we call the dark matter. So itís a whole different technique in clusters for looking at the amount of dark matter thatís present as opposed to, say, the rotational curves in spiral galaxies. And actually, it turns out, elliptical galaxies were discovered to have X-ray halos, extended gaseous emission. You could measure the gravitational potential in elliptical galaxies using the same kind of technique.
McCray:And then compare that with optical?
Tananbaum:The light to mass ratio from the stars can determine the dark matter in ellipticals. So can the distribution of the X-ray gas, so that was a very significant capability that was started with the Einstein.
McCray:The timing of this launch, I donít suppose it was done on purpose, but its mission ended around the time that George Fieldís National Academy of Sciences decadal survey was going on. I wanted to know if you could say something about how you helped sort of lay the path for Chandra versus those in the community who wanted large optical, or infrared, or ultraviolet telescopes.
Tananbaum:I actually donít have too much information to share in that area because I wasnít actually on the committee or the subcommittees for the Field decadal survey. I served on a panel for the 1990s. But I was only vaguely, letís say, aware of the various discussions that occurred. There is no question that the results that came back from Einstein and the kinds of things that one can do generated tremendous support for a longer lived, more capable X-ray telescope, which is what AXAF/Chandra are. The fact that we had already started studies on the mission were also very, very helpful. The fact that probably IRAS [Infrared Astronomy Satellite], the first Infrared Satellite, hadnít yet flown gave us a certain advantage because IRAS was a tremendously successful mission.
McCray:1983, that was launched.
Tananbaum:And in fact, it encouraged some of the infrared community to think about whether SIRTF [Space Infrared Telescope Facility] should move ahead of AXAF. Well, SIRTF had been envisioned in 1980 as flying on a shuttle for a week or two at a time. And then we began to realize the cost of the telescope and that it would only have this limited amount of use if it was flying on a shuttle. And the science potential once IRAS flew, of course, it was tremendous in the infrared band as well. It was clear that the idea should be to make SIRTF a free flying observatory. And then the whole question of how you get AXAF and SIRTF going, particularly given some of the difficulties Hubble was encountering, led Charlie Pellerin, who was the head of the astrophysics program at NASA headquarters, to convene a group of scientists, Iíd say in the 1984, 1985 time frame. Martin Harwit was asked to chair the group. I was one of the people invited to the meeting. And out of that meeting, the concept of a Great Observatories, erogram, I think, was first stated at the meeting. The idea that you could use these observatories in concert along with the gamma ray observatory which, of course, flew and finished its mission before either Chandra or SIRTF got up. But in any case, the overlap of the observatories, the multi-wavelength studies that they would enable, and trying to now explain them as an ensemble with connectedness as opposed to discrete, competing individual observatories was a tremendous advance to help make the Field report a reality. And then the Bahcall report for the 1990s actually recommended SIRTF as its top priority.
McCray:Well, since you were involved with the Bahcall report, were you on the main panel for that?
Tananbaum:No. I was on the High Energy panel.
McCray:What did you see as part of your participation on that subpanel?
Tananbaum:It was interesting in that the GRO was about ready to launch or had just launched as the panel was finishing. But AXAF was still in this state where we had built the largest pair of optics and demonstrated that they could be done to the satisfaction of the Congress, which was this big challenge. But then it was decided that there still wasnít enough money to go ahead with the flight program. The program had to be restructured. Well actually, the restructuring occurred in 1992. So while the Bahcall committee was meeting in the 1990 time frame, it was very important that we somehow re-endorse what the Field Committee had said and say that doing AXAF was still our top priority. But then also make some room for some new initiatives. It was pretty clear that it would be very difficult to put another X-ray mission at a very high level in the same box. So the X-ray community was supportive of what became lower cost MIDEX and SMEX Explorer opportunities.
McCray:What are those?
Tananbaum:MIDEX are Medium Explorers and SMEX are Small Explorers. So there would be opportunities to possibly do something that one of the big telescopes wasnít willing or able to do, but on a smaller platform. We were generally supportive of the SIRTF mission. One of the agreements we made through the late 1980s, so that we could get the whole community working together was that SIRTF ought to have very high priority in the next set of reports following AXAF. The SIRTF people agreed that after a certain amount of give-and-take that trying to turn the order of things around would end up being very harmful to all involved. So support of AXAF going ahead was agreed on.
McCray:In 1980 decadal survey?
Tananbaum:Well no, not in the survey in the 1980s, but in the later 1980s when there was this pressure that SIRTF should be a free flyer. Then should the Field Committee be reconvened, for example, and reprioritize between SIRTF and AXAF? And that was seriously discussed and it was decided that it would be counterproductive. AXAF had already been under study as a free flying mission and it was closer to being ready to build, and if the community kept changing its priorities, it would have limited or no credibility. And so, part of this Great Observatories Group and some smaller groups of people had met with Charlie Pellerin. The rationale was good and, of course, easier for us to accept that we would continue being the next big astrophysics mission. But in return, we needed to make sure we all supported SIRTF. So we certainly did that through the Bahcall Committee process and afterwards I was part of several of the NASA advisory groups in the early 1990s. These, I think, contributed to some of the strategies to help get SIRTF so that if we launch SIRTF as planned now in early 2003, it still would be about two-and-a-half years after Chandra launched in mid-1999. But not ten years, which is the Hubble-Chandra difference. In any case, for the High Energy and for the X-ray community the decadal survey for the 1990s was this fairly tough situation of making sure we just retained the priority that weíve gotten for the 1980s. And somehow also help to generate a new program for the 1990s.
McCray:Of the Great Observatories, it has always struck me that SIRTF has sort of had the worst luck. When it is launched next year, NGST [Next Generation Space Telescope, re-named James Webb Space Telescope in September 2002.] is sort of following hard on its heels if it is launched when NASA says it will be. I have been kind of curious about it being such a relatively modest-sized space telescope being followed maybe seven or eight years later with something that is an order of magnitude bigger. I was curious if you had any thoughts on that?
I donít know that SIRTF has had bad luck anymore so than Hubble or Chandra. If you look at the timeframes from which things were first suggested, Lyman Spitzer actually wrote a paper for one of the classified agencies; I donít know if it was the Defense Department or reconnaissance. But he wrote a paper in the late 1940s about possible telescopes in space and realized that you could do a lot of interesting astronomy with a space telescope. The space telescope people starting actually working on it in a serious way in the 1970s and it launched in 1990. For Chandra, if you start with 1963 and Giacconi you have 36 years, or if you start in 1976 when we sent in the proposal and I actually personally started working on Chandra—from the time we proposed and started working on the technology until it launched—you have twenty-three years. Obviously, there are a few years more since then. It is a lot more fun now that itís flying. So SIRTF was around from the late 1970s or early 1980s; so it is 20-25 years. It is not the way you want to see things and you would like to have a great concept, study it, build it, launch it. You would like to see all that over a ten or fifteen year time frame perhaps.
The only thing Iím saying is I donít think SIRTFís luck, as it were, or the series of steps that it has had to go through was all that different from Hubble or Chandra. Now from the point of view of NGST, if it launched by the year 2010 that would, of course, be a tremendous success because the technology challenges are immense. While a lot of progress is being made, it will certainly have its challenges as it is being built. It can only break the budget up to some point and so it will either be technology or funding constrained. If somebody play this tape back after 2010, weíll see if it has been launched! But even that said SIRTF is a tremendous advance for the infrared community. It will return this wealth of data that people will just, for roughly a decade, be the best data we have in the infrared band. And will complement things from Hubble, from Chandra. You know, there are AGNs we see in the X-rays. They donít seem to give any hint from their colors or even their optical spectra of any kind of an accretion disk around a black hole in the center. There is just no signature of activity in the optical. Then you go into the redder part of the optical, this recent All Sky Survey called 2MASS, the 2 Micron All Sky Survey, found all sorts of red galaxies, and red stellar objects, and it has found a new class of red quasars. Some of them have broad lines; some of them have narrow lines. But they are objects which we hadnít been picking out with the classical techniques in the optical.
There either is so much absorbing material around the accretion disk, around the place where the X-rays are being made that most of the optical light doesnít get out, most of the lines are smeared out. Or some other process which weíre not exactly on top of right now could be occurring. The infrared is the place where a lot of that absorbed luminosity could get re-radiated. So it will be incredibly exciting to look at the X-ray, optical, and infrared properties of a set of AGNs, for example. And SIRTF absolutely has the sensitivity to do that. Now eventually, if you want to do it at the redshift of 10, then maybe we need NGST and probably even post-Chandra as the X-ray part to do that. But there is this whole wealth of star formation problems, and dusty galaxies, and AGNs that are absorbed. I think SIRTF is going to be a tremendous success from the science perspective. And from a scientistís point of view, in this case it has definitely a finite life because of the refrigerant that cools the telescope, even if things go well, I think about five years is the most one could reasonably expect to get. So having another telescope capable of doing even better and further in the infrared, you know, a few years afterwards. We shouldnít have these gaps of a decade between one X-ray mission and the next or one optical telescope or infrared telescope and the next. You have all sorts of impacts when you have those kinds of large gaps. Where do the people who build instruments and optics go in the interim period? Why would students want to work in a research area where there isnít a telescope available to take new data? So how do you shape, and train, and involve the next generation of astronomers in a particular discipline if you put these large gaps in?
McCray:This sounds as if it is a serious problem for the astronomy community.
Yes. I think it is in general. I think that the ideas which have been embodied now in the NASA roadmaps for the future is to not just look at a single mission, but to look at a plan over a twenty year period. These roadmaps and strategic plans, involve missions plus working on the technology for the longer lead items; for example, those needed for an X-ray interferometer downstream. Weíre working on the Constellation X-ray Mission. Itís now the successor to Chandra. Itís complementary in that it emphasizes spectroscopy. Right now, if we can get into the NASA budget and start building it around 2006, we could fly it by about 2010 or 2011. By the way, that will represent a fifteen-year program. The proposals that led to the current Constellation concept came from a group that we formed with myself as the Principal Investigator and another group that was led by Nick White at Goddard Space Flight Center.
Those two proposals were selected for study around 1995. We put the two of them together as an integrated team in 1996 or 1997. But it will be fifteen years from the time we started talking about such a mission until we fly. Things are going very well. But we could get a big budget disruption and not get the funding in 2006 or 2007 for a variety of reasons. And we are not funded at a level such that all of the people who could develop optics or instruments can be supported on Constellation quite yet. Having this plan is helpful from the point of view of seeing what weíre doing post-Chandra and then we are planning something called Generation X, which is a 100-square meter telescope with subarc second resolution. For X-ray astronomy. And something called MAXIM which is an X-ray interferometer to do microarc second imaging to actually take an image of the material outside the event horizon of nearby black holes, for example. These are what NASA and Dan Goldin [NASA Administrator until 2001] call ďvision missionsĒ. You might envision that the vision missions will probably end up being significantly different from what theyíre currently conceived as. But having some sense of doing these within the context of a program, I think, helps to give a picture to students and post docs that if they work in this field, what might be happening, what they might be in a position to influence, and participate, and actually do science. It certainly helps us from the point of view trying to focus some of the technology investments. So this is a little bit of change from the 1970s and 1980s where sometimes we were just talking about one mission. And then when it was done, start to talk about the next mission.
McCray:So this is more of a continuous way of looking at it?
Tananbaum:Right. And the planets, the strategy for Mars and the outer planets has been developed in a similar way at NASA. And certainly the set of Astronomical Search for Origins programs consisting of the SIM [Space Interferometry Mission], NGST [Next Generation Spack Telescope; now named JWST for James Webb Space Telescope] and, Terrestrial Planet Finder are connected in those ways. Although I think one would argue that the science of NGST, for which it is most advocated, is quite a bit different from the Terrestrial Planet Finder science. But that said, those telescopes will be used for some very similar kinds of science and have a concept for the field overall. And then, of course, folding in the ground-based optical and radio and I expect gravitational waves, neutrino searches. I mean, there is a lot of stuff that all falls together for what we call the astronomy and astrophysics for the next twenty years.
McCray:That leads nicely to a series of questions about the physics and astronomy connection, but I need to take a break for just a moment. [Tape cut] I am curious on your thoughts today about how distinct physics is from astronomy as a separate area of research. I am asking this also knowing that you were on a recent NRC ďThe Quarks to the CosmosĒ panel.
I donít think that there is a huge separation from astronomy and physics, or a separation between them. At some level, astronomy is a physical science. It is done a little bit different from the way particle physics is done. It is done a little bit different from the way condensed matter physics is done. You canít control the experiment in most cases. You have to choose more of what you are going to look at. Weíve talked before about an accelerator experiment, when you smash a couple of things together, you have some reasonable expectation of what kinds of particles might be made. But the reason youíre doing the experiment is to see what comes out of that explosion youíve created. The only thing youíve done there is youíve chosen to smash particle A into particle B. In the case of astronomy, weíve got to use natureís accelerators to do that and figure out where they might be. The kinds of skills that one needs to be successful in astronomy today have a close connection with what you need to do in physics as well. You have to understand statistics to be able to detect a signal whether itís in astronomy or physics. I canít imagine much being done these days without a computer to either control the experiment or analyze the data.
So a knowledge of software techniques. The detectors that we use in X-ray astronomy, and I think in other areas of astronomy as well these days, have a common heritage to detectors used in physics labs. And certainly, one is always looking at them for medical applications and other commercial applications nowadays. So I think a lot of the techniques are similar. If you wanted to get into the details of the radiation processes that are generating the X-rays that we see, whether itís the continuum parts of the process, the scattered, or the lines and absorption features, you get into the same basic areas of atomic transitions and quantum effects. So you need to have a grounding in physics if youíre going to do that kind of work. I think that everybody becomes a bit specialized at some point, either in graduate school or as a post doc. So people, whether you call them astronomers or astrophysicists, there may be a difference if their degree is in physics or it is in astronomy. But, in these times, a good astronomer will have to have a solid grounding in physics as well.
McCray:Do you think that has become even more so since you went to graduate school?
Tananbaum:Well, my graduate training at MIT was a physics program. So I probably donít have real good basis for comparison other than to realize that over the years I have written letters of recommendation for various people. And in some cases, I write to the physics department. And one of the things I say is that having a couple of people who are very active in astrophysics should be a very real element of your program, if it is a well-rounded physics program. Sometimes I have written to astronomy departments telling them that they need people working in high energy astrophysics or X-ray astronomy because that is a critical element of what is happening in astronomy these days. So I think, yes, that departments and universities in general have tried to, one way or another, incorporate astronomy as part of an astronomy and astrophysics and physics program and high energy astrophysics is part of that. Probably there has been a narrowing of any separation that might have existed between astronomy and physics as disciplines.
McCray:How much intellectual interaction do you think there is between people who observe in, for lack of a better word, the traditional part of the electromagnetic spectrum and the people who are doing work at the extreme ends of the spectrum, or the gravitational wave communities?
I think there is quite a bit of intellectual and actual collaboration between the people. It can vary. An individual could choose to work on a particular calculation, or model, or source and do it in whatever way he or she wants. An individual could choose to, if they wanted to, study Eta Carinae and if they wanted to study it in the infrared, the optical, and the X-ray, they could do all of those and not interact with another X-ray astronomer, or another infrared astronomer, or another optical astronomer. They could choose to be knowledgeable in these areas. And there are people who do things like that. I do know that on some of the projects that we have worked on in the past and that we are working on at the present, getting optical identifications for X-ray sources is a very critical component to knowing what kind of an object it is, how luminous it is, how far away it is. What you need to do to get at the underlying physical processes usually requires optical, and often radio, and infrared data in addition to the X-ray. And we find it efficient, I think is as good a word as any, if we team with somebody who knows the techniques of optical observing, knows the best way to get the telescope time and to make the observations.
We have some of that expertise in-house within the high energy division here. We have some within the Center for Astrophysics. But there is a very much an openness to collaborate with people at other institutions. Obviously, if you are going to publish a paper that involves optical data as well as X-ray data and you want to put your name on it, you ought to have some understanding of whether the optical data, the measurements, have been obtained with the same care that you would want for the X-ray. And whether the data have been reduced and analyzed in a sound way. Of course to some extent youíre trusting your colleagues and co-authors to do that. You at least look at it and see if it makes sense to you as one of the co-authors on such a paper. But then, there is an efficiency of drawing on the skills of experts in different areas and the synthesis of that expertise to then come out with a result for a project, or an object, or a set of objects. I think that much of what goes on in astronomy, most of it, is much more problem oriented rather than wavelength oriented. The Great Observatories concept has contributed to that. The fact is that with Chandra the reason we have the Chandra X-ray Center is so that each individual doesnít have to figure out exactly how to program the telescope to do the things it needs to do. We have modes to do a particular observation the best way, and we established the sequence of the observations, and we generated some standard software pipelines to produce standard images or spectra. We donít want each individual scientist to have to invent that just as each one wonít have to build his own telescope or her own detectors.
And Hubble does the same thing at Space Telescope Institute and the SIRTF Science Center will be doing the same thing for SIRTF. So we try to make it easier for an individual who hasnít necessarily worked in the X-ray band or in one of these other bands to have an idea that, ďGee, getting X-ray data for such-and-such an object or group of objects would really enhance my science and really make a lot of progress.Ē They donít have to be an expert at X-ray astronomy in order to write a good proposal to get it through the peer review and get some observing time. And then there are also these tools and these standard techniques to at least get started on the data analysis. We provide a helpdesk, we provide some additional support if people run into trouble. So I think that NASA and I think to some extent, of course, the national ground-based observatories as well have been setup to enable and at least facilitate observers to think from an object oriented and a science-driven perspective as opposed to a wavelength-driven approach to the science. And it is clearly a better way to do things. By having these centers, I think it keeps the learning curve from being infinitely steep or having ten steep learning curves you have to go up and avoids it from being very inefficient in terms of time and certainly duplicative in terms of cost.
McCray:Do you have any anecdotes or comments about this recent NRC report, The Quarks to the Cosmos? I realize it just came out.
I participated in the second stage, so some of the phrasing of the key science questions had already been worked in the first stage. And in the second stage there were presentations from various programs, and missions, and we tried to make assessments of the technical readiness and where you could make progress on addressing this group of questions that was defined to be at the intersection of astronomy and physics. And so that means that there are things that you might do which are primarily of interest to an astronomer or things you might do which are primarily of interest to a physicist. The focus of some of this study was to identify things which would be of interest to both astronomers and physicists. And to think about ways that either an astronomer, or a physicist, or perhaps a scientist with a relevant mix of skills could address those kinds of questions.
So I think it was probably very useful in terms of generating support for some of these things where astronomers might say, ďWell gee, why doesnít the physics community do that?Ē Which really means, ďWhy donít they come up with the money?Ē to do that. Or the physics community would say, ďWell gee, thatís mostly astronomy. Why donít the astronomers pay for it?Ē They tried to put it out in an area (the overlap between physics and astronomy) and begin to think about processes by which NSF, DOE and NASA might collaborate on some of these programs. It is always a problem because funds are limited and if something is of interest primarily to one discipline, or one area, or something that falls in the middle, itís almost like a double jeopardy. You have to get two sets of approvals. And anything that is almost infinitely hard, it becomes infinitely squared hard. It is very, very difficult. So hopefully this study will enable some of these important research areas to obtain support. Of course one of the reasons that people are interested and willing to participate in these kinds of studies is the hope that it will be an opportunity to actually increase the funding for the field. And not so much that weíll do A, B, C instead of D, E, F which were previously planned because that just leads to these turf wars and the like. But setting something out there which is really fundamental, which is really exciting from a scientific perspective.
Letís say youíre at a point in time where you have or almost have the expertise, the tools to build the telescope, or to do an underground lab, or in some other way facilitate the thing. Once we know the lifetime of the proton, then we know the lifetime of the proton. Okay. Or the mass of the neutrino. Theyíre very important, but once theyíre done, itís also true that theyíre sort of done. So itís an incredibly exciting time. We know thereís a lot of dark matter out there. We donít know exactly what it is but weíre beginning to get a better and better handle on how much of it. Now all of a sudden, dark energy comes along and changes the whole view of the expansion of the universe. How are we going to determine what it is, how much of it is there, or what does it really tell us about cosmology? I think it is mind-boggling. I mean, if you had asked anybody five years ago to tell you what dark energy might be, maybe there was one theorist who had worked on one set of equations. And certainly the cosmological constant in Einsteinís equations can be used to be the equivalent of the dark energy in some ways. It certainly can be used to explain the change in the rate of expansion of the universe. So whether it was his biggest blunder or not, whether it should or shouldnít have been there, notwithstanding it is now interesting.
When Vera Rubin first talked about dark matter, which is probably twenty-five years ago give or take, people thought her data just werenít any good. Redo the observations, reanalyze the data, look at more objects because it was such a radical concept. Now you show somebody either, from rotation curves or from the gas not escaping from the halo of an elliptical or from a cluster, that thereís a lot more gravitational pull there holding the system together or making it move the way it goes. You write the equations down. Just about everybody believes, yes, thereís more matter there. Then you can get into a debate, is it baryonic or non-baryonic. Is it really dark and so on and so forth? Now weíre just on the threshold of that happening with dark energy. So I think the fact that astronomers are interested, physicists are interested, a cosmologist is he or she a physicist or an astronomer? Or is he or she a cosmologist? I mean, what difference does it make? What matters is to articulate these questions and to generate enthusiasm for ways in which we could begin to answer them.
McCray:It sounds as if youíre describing a situation where a lot of these more traditional boundaries have dissolved or become more permeable between cosmologists, physicists, and astronomers.
Tananbaum:Thatís probably true. I usually say Iím a physicist when people ask what I do. Probably because of the fact that the Ph.D. is in physics, and certainly I think there was more prestige thirty years or forty years ago in being called a physicist than an astronomer in some general way. But I interact with and have a tremendous respect for many of the people that are in the astronomy community. I donít go and say, ďIs your degree in physics or is your degree in astronomy?Ē Rather: What are your interests? What are you publishing? How does that interact with the things that weíre interested in? These are the important questions. So I think the distinctions are almost immaterial. They may be very important as you apply for a job at a particular university. If your degree is in astronomy and itís in the physics department, maybe thereís still some extra threshold that one must pass. But what is really relevant is the work that one has done. Is this work, whether itís experimental, whether itís observational, whether itís theoretical, is it raising the level of our knowledge in some important area? Is it an important contribution? Does it show original thinking? Does it show carefulness? Do you treat the data with respect? Some people have wild ideas and putting wild ideas out there is a stimulating and a useful thing. But then somebody who does a measurement that checks the wild idea canít just do a wild experiment. They have to do it carefully.
McCray:You had mentioned at the very beginning of our interview that you grew up in a fairly religious family, if I understood right.
I donít think it was particularly religious in the sense that we didnít go to services every Saturday. I didnít go to services every morning and evening. But it was probably slightly above the average in the sense that I went to Hebrew School (in addition to public school) and continued Hebrew studies beyond my bar mitzvah through high school. I actually had a scholarship. A couple of years before I went to Boy Scout camp which was a very different kind of experience, I went to a Hebrew speaking camp which was an immersion kind of experience. And I probably was the only kid who didnít speak very much Hebrew the entire two summers that I was there. Not that I couldnít, it was more that I wouldnít. And certainly on the high holy days, I still take off for Rosh Hashanah and Yom Kippur as a family. Now the kids are away, but as a family we tended to go to services. I think that for me itís partly a sense of community, itís partly a sense of the holidays as a useful time to just reflect how one is living oneís life, what one is doing with oneís time professionally, personally.
Thinking a little bit about, maybe itís a little corny, but being a better person. I think the value of the religion to me starts almost with the Golden Rule concept in the sense that our measure, as near as I can figure it out, is based on what we do while weíre here. I donít have any hard evidence for there being anything after weíre here. I have a lot of evidence for things we donít understand as scientists. But I think from the point of view of religion, staying with that first, and it doesnít have to be through religion, but for whatever reason it gives you a sense of the importance of caring about other people and caring for people. Essentially I have said to people here when they have come in and they have been frustrated with a younger staff member and not quite sure how to either nurture or boot them out the door. I have often said, ďWell imagine that things were different and it was your young adult child that was working someplace. How would you want that place to treat your child?Ē Well, thatís just the Golden Rule again, right? It is interesting to put it in a perspective. ďWell, I would want somebody to take some time. Iíd want them to work things through. Iíd want them to figure out eventually if this is the right place, the right role. And if not, maybe help them to find a better direction.Ē You get a lot back when you help people to be successful.
You get their friendship and good will, but you also get this tremendous loyalty and commitment to the success of the project. So a lot of what we do, weíre a big group. And so a lot of what we do, we do build things, but what is unique and what is special is the collection of intellectual skills that our team brings to the projects. And so, you want to have an environment in which people are encouraged and enabled. When they do have a problem that there is someone that can take care of it (or try to help), whether itís professional or personal. I think some of that sense has contributed to some of the success weíve had. I should know, but I donít, we have roughly sixty Ph.D.s on the science staff of the Chandra X-ray Center. There are more, other Ph.D.s in high energy astrophysics here. It is almost three years since Chandra launched and of those sixty Ph.D.s between here and MIT, I believe only two have left in the three years since launch.
McCray:Thatís a pretty good retention record.
Yes. Some are being recruited, some probably arenít being recruited; some are probably doing fantastic, some are probably just doing okay. But we have a situation which says that itís a pretty good place to work because people are staying. There are jobs out there that they arenít running off to those jobs. I think that weíve had some success. I donít even know precisely the numbers, but pretty good success in recruiting and retaining the women scientists on the staff. One reason is that you always run up to this question of families. Somebody gets pregnant and obviously they need some time. We say they can work from home, they can work a three-day week, they can work one fourteen-hour day. Weíll work a job to whatever the individual feels, she, and then sometimes in the case of a spouse, he, wants.
I know all the little babies just about, theyíve all been here when the daycare fell through on a critical day or when summer vacation starts. I love children. Itís nice to have seen some of the children that I remember as babies that are now college graduates. So it is not that itís one huge family; with the number of staff we have, it clearly isnít. But we basically have had a lot of flexibility, so thatís a value to family which comes again from some of the early upbringing and family experiences. The value of education. We hire kids with—Iím supposed to not use the word ďkidsĒ—younger people with a bachelor degree in science who arenít quite sure if they want to get a Ph.D. or what direction they might go. We have, probably, twenty-five or thirty such people on the CXC, Chandra X-ray Center, staff. In any given year, a half-dozen to a dozen may go back to graduate school after having worked for a year, or two, or three. They get a sense of excitement about what they might be able to do. Maybe itís X-ray astronomy, maybe itís some other area of astronomy. Theyíve done some real work. Theyíve put their name on a couple of papers for which they helped with the analysis of the data, and the observations, and the like. I think itís a wonderful program. Itís sad sometimes to see them leave because theyíre fun to have around. But I also know that theyíre embarking on the next step of their education, or lifeís adventure. And again, the value of education is something that, again, dates back to how Jewish religion puts a high value on education. Certainly education was an opportunity for members of my own family too.
McCray:I never asked you about your siblings. Did your siblings go on to university?
Tananbaum:Yes. I have a sister and a brother. My sister has a bachelorís degree, taught at the high school level for a number of years and runs a housing agency for the elderly in Buffalo. And my younger brother has a Ph.D. in history and teaches on the faculty at Lehmann College in New York. I also have 2 grandsons (Kyle and Jason) in Dallas and one more grandchild on the way in California. In any case, going back to the earlier statement, I think from my perspective the family and to some extent the religion are intertwined. It is a sense of a community. It is a sense of roots at some level even if those roots trace back to some place in Eastern Europe and then sort of disappear. One doesnít get up one morning and simply say, ďThese are my values.Ē Those are things which sort of get in through the pores, and into the memories, and into the brain. So that is whatís probably, from the point of view of values and professional life and the impact as a person, those are the things that I would say. I donít know if that makes me deeply religious, or agnostic, or something in between, but that is probably how that component of my upbringing and my life have been affected.
McCray:I had one last question and it is the most open-ended question saved for last, of course. Over the course of your career, what do you see as the most significant change that has happened?
Well, I think that over the course of the career the changes, you would have to separately consider personal and then the development of the field. Of course, the development of the field is in some ways easier to talk to partly because it is less personal. So from the point of view of the development of the field, it has clearly gone from the study of a few bizarre and completely not understood objects, to a powerful telescope able to take pictures in the X-ray band that are comparable to the ground-based optical and radio telescopes. Weíve studied almost any and every type of object in the universe and study it in ways where now we are looking at the potential for understanding the physics of the black holes, or the dark matter, or the size and age of the universe. And attacking some of those global questions from a somewhat different perspective. I think it is important, for example, for dark energy to not just have it all depend on the supernovae and the distance scale that one gets.
With the Sunyaey-Zeldovich (X-ray and radio) measurements, one has at least the potential for independently determining distances to clusters of galaxies. And from redshifts comparing the distance that one would infer with different cosmologies and seeing what the distance scale tells you as measured by that technique. Itís been a little disappointing with some of the uncertainties that we found due to the structure within some of the clusters and changes in the instruments, the CCDs got a little bit of radiation damage early on. So while we had hoped to calibrate and do things at the 1% or 2% level, weíre still struggling at more like the 10% percent level. I think weíve got data in hand that I think we will eventually make a substantial contribution in that area, but itís three years and no papers to speak of so far.
So I would say for the development of the field, the techniques evolving from the collimated proportional counters to arc second telescopes, and how we are talking about the future technologies for microarc second X-ray interferometers. Weíre talking going from a few square centimeters for the earliest detectors to 100-square meter in Generation X telescope size. Weíre talking about spectrometers with 1-eV resolution as opposed to proportional counters with 1,000-eV energy resolution. So those are thousand-fold increases in terms of the capabilities or more. And certainly the numbers of people. There were probably two dozen X-ray astronomers when I was in graduate school.
McCray:You probably knew them all by name.
Yes, just about. And we have hundreds of people, as Principal Investigators of general observer Chandra proposals and thousands of people making use of Chandra. So there has been this very substantial increase in capabilities, an increase in the terms of the kinds of science that is being done. To be in the middle of a lot of that is, of course, very satisfying from a personal point of view.
From a personal perspective, at twenty-five youíre in graduate school and youíre not even exactly sure what youíre going to do. Now Iím a few years older than twenty-five. Iíll be sixty in another month. I am still very enthusiastic about what I do. I hope to continue doing it for a number of years until they roll the cart up and say, ďWe need your office.Ē
I wouldnít have imagined myself working on a large project with a large number of people, managing an effort, going to Washington and talking to people about why they should fund it when I was twenty-five or even at thirty-five. It has been an opportunity for me and I do as much management or more than I do technology or science. I think, actually, it has turned out to be a wonderful mix because there are many better scientists here who take on particular technical and scientific challenges.
We laughed earlier about the engineers donít want me around the hardware with a screwdriver in my hand. My former thesis advisor, Jim Overbeck who no longer works in the field, I think shook his head in amazement when he heard Giacconi was leaving and that I was going to become the head of the group. It was about fifteen years after my Ph.D.; about twenty years ago. I donít think he envisioned that being the direction that I might necessarily go.
I think there are people outside of the group who expected when Riccardo left that within a year or two, everybody would go off in a different direction. It never occurred to me that we wouldnít succeed even when Riccardo left because Riccardo had, in a sense, trained me and some of the other senior team members very, very well. Not just in the technical aspect of things, but taking us to different meetings where things were discussed with NASA. My seeing them actually turndown a request to add the magnetic torquers for the Einstein satellite and then seeing the strategy of our writing a proposal and in a sense seizing the opportunity. If we got turned down on one thing, then we didnít just sort of crawl back into our cave and lick our wounds. We figured out how to then make the field go forward, to be success driven.
I think then that I knew that the people who worked here were used to working with me in a leadership role and would be very confident that we were going to continue to be very successful, which we have been. So I think it has been very rewarding from a personal point of view to sort of look back at that evolution from an individual scientist to a leader of a team of people. I think it has been very rewardingĖ relationships with the people we work with at NASA, with industry, people up on the Hill.
I have formed a lot of very, very good friendships, people that I worked with who were in a position to help us that interacting with them, I think theyíre friends as well. Dick Malow would be a case. I donít know him away from work in a deep, deep personal way, but we stayed in touch, and when we see each other heíll ask how AXAF/Chandra is going. We talk about the fact that the challenge that he gave us (to demonstrate that the first pair of large Chandra mirrors could be built) helped to meld the team into a real team; that was a trial by fire for NASA, industry, and ourselves. We had to succeed and we had to work together in order to succeed. I donít know that that was Dickís intent. His intent was that he didnít think we could do it and then if we couldnít do it, he wouldnít have to come up with the money for the project. That was probably his intent. But at least he gave us an opportunity to show that we could do it. I think that I am very fortunate. I have friends who are successful in lots of different venues and financially probably in some cases make a lot more money.
But work for them sometimes is really just sort of work and when they see me working at night or on a weekend or they hear my wife saying, ďWell, heís doing this, and this, and this so weíre going to be a little bit late.Ē They actually are able to sense the excitement that I have and that most of us who work in astronomy, I think have. But Iím just talking about me right now. The excitement that I have for what I do. I did complain earlier this morning that Iím trying to figure out whether this particular request for Directorís Discretionary Time is a good or a bad thing to do. Will it work or wonít it work? But my mindís going a hundred miles an hour learning a little bit more about that area of science, learning a little bit more about what our instrument could or couldnít do. Thinking about if we were do it, can I help the person who proposed it to even do it in a better way? I have fun with that. I mean, there are frustrations. There are nights when I go home that Iím tired, my project didnít get funding. There were times when an intelligent person would have quit and gone and done something else. Iím also stubborn. But the fact is, is that Iím one of probably a small set of people that is very, very fortunate to be discovering things that have never been discovered before and I get paid to do that, right?
When I talk to teachers or I talk to a general audience about what we do and I can see their excitement and enthusiasm and hear, ďGee, the pictures are beautiful.Ē And the things that weíre doing are so exciting. So itís just a wonderful thing to be able to help make some of that happen and to be so excited, enthusiastic about what I do. That is true just about everyday. I can remember when the kids were little when I coached their Little League baseball team, I had to be at the game at three oíclock. I had to make sure I got out on time and got there. My younger son Greg in particular says I was very, very good about doing that and being there for the kids when there were things that they wanted to do or we were going to do together.
The fact of the matter is other than that, I can probably think of two days in the last thirty-plus years where the day seemed to be dragging and I couldnít wait for it to be five oíclock because it would be quitting time. Every other day at some point I have to say, ďMy goodness, I have to stop what Iím doing now,Ē because it could be ten oíclock and I have to either go eat something. Or itís eight oíclock and I need to get out and run before it rains or it gets too dark or cold. Thatís great. I can login now because with computers I can login from home at midnight or at 7:00 AM and just check if everything is all right, if everything has been quiet.
McCray:Do you check often?
Tananbaum:I check late at night before I turn in. I check in the morning before I head to the office because if something unexpected has come up, I either can use the cell phone coming into the office so I can at least think about what are some of the issues, how do we want to proceed, whatís the risk, whatís the gain. By the time Iím at the office, I canít type as fast as my mind has been working. Already, Iíve jumped through a half-dozen hoops. On the weekends, it depends on where we are and what weíre doing. But I typically check a few times over the weekend. Things are running so smoothly and weíre in sort of a stable part of the mission, so email traffic can often wait other than requests for unexpected targets that can come in over the weekend. Email traffic is generally pretty light and thatís fine. Itís good to be able to get away for a weekend.
Well, I think we should probably close there. Thank you very much.
Session I | Session II