Wolfgang Panofsky

Lynch:

This is an interview with Wolfgang Kurt Hermann Panofsky done in his office at the Stanford Linear Accelerator Center on 11 April 1997 as part of the AIP oral history program. The previous interview in 1974 dealt with the construction of SLAC but it did not discuss the ultimate product of SLAC as it was intended, namely to do physics. Let's discuss the various phases of the physics, its planning, execution, and the surprises. The cornerstone of the original concept of SLAC physics was the set of experiments done in End Station A. There were 3 spectrometers built there. How was this complement of spectrometers chosen and to what extent was the choice an outgrowth of spontaneous proposals as opposed to a grand design?

Panofsky:

It was not a grand design. In fact those 3 spectrometers were chosen by groups. First a collaboration headed by Dick Taylor, in which I was a member, chose the first 2 spectrometers and the third spectrometer was independently proposed by Dave Ritson and his collaborators, and it was simply added. In the original proposal the idea that there shall be spectrometers was already included, but if one looks at the original proposal the concept introduced there was extremely naive and rather primitive. One of the immediate things in which I was personally involved was to decide that the optics was wrong which we inherited, namely point-to-point focusing, and that in fact the right way to go is to have line-to-point focusing because liquid hydrogen or liquid deuterium targets traversed by the high-energy electron beams would be a relatively long source and therefore one would gain in intensity by having that line source focused onto a point and then the different angles of production would map into different points.

Therefore in the final focal plane there would be a display of angle of production in one direction and energy dispersion in the other direction. So that feature, that basic optical characteristic, is common to all three of these. But the actual design was fairly well separated and the 3 spectrometers covered different energy and angle ranges. And in fact, in the original proposal, there is a picture, sort of a joke at this point, which shows a single spectrometer sweeping the full 180 degrees from the forward direction. Kinematically this doesn't make any sense because of the fact that the secondary energies decrease very rapidly as you go the backward direction. So therefore the decision was made to have firstly two spectrometers, one which would cover the forward direction with instruments going to the highest energy but relatively small solid angle, then the second one of medium solid angle covering the intermediate angle range, and then finally Dave Ritson proposed a relatively simple but quite sophisticated optical design to cover the backwards angle. So, this turned out to be a big engineering project and Dick Taylor was more or less in charge of this; I stayed directly involved but only in the first experiment, which covered elastic electron scattering, and after that I phased out of the direct collaboration with this and the very able people in Group A carried on.

Lynch:

At what point did you realize that this one spectrometer doing 180 degrees was not a good strategy?

Panofsky:

I don't remember, quite early. I mean that was done sort of in a heat of passion during the original proposal writing stage, and on thinking about it I recognize that this was not the way to go; I don't remember when, but soon thereafter. Also, there is a famous cartoon, which I don't know how it was made, which was a vertical spectrometer covering the full energy range which was sometimes used during presentation of the original proposal where the top end was covered in clouds. It's a nice cartoon; we have it around somewhere.

Lynch:

That sounds like a twist on Bob Hofstadter's machine down at Mark III.

Panofsky:

That's right. That, needless to say, was never taken seriously by anybody, including Bob Hofstadter.

Lynch:

What were the interactions like between the 3 groups that built the spectrometers?

Panofsky:

As far as I remember, reasonably peaceful. I don't think it was a competitive thing at all. One thing which we decided, they were totally complementary in coverage; this we recognized and they had common basic optical features. The collaborations were different. The electronics in the 20 GeV and 8 GeV spectrometers were largely done by the MIT collaborators while the magnetic design was done here at SLAC and the contracting out of the magnets and all that was done here at SLAC.

Lynch:

Was there cross-fertilization between the two?

Panofsky:

Again, the details I don't remember. I recall it was quite peaceful and we had large enough budget so it was not a budgetary problem. The whole atmosphere at that time when we started the research program was really quite different. We provided for a rather generous equipment budget to cover all the major pieces of equipment. One main thing concerned me from the very beginning: I was very much impressed by the fact that, because of the bad duty cycle of the machine, we needed major spectrometers and major instruments. Building block experiments would not work because of the duty cycle. Therefore, based on that principle, I was able to persuade AEC fairly early in the game to establish a rather large initial equipment budget, which was much larger than what was customary for proton machines during that time. Therefore, there was, I believe $18 million of then-year dollars, which was a lot of money, and therefore there were funds available for the 3 spectrometers as well as streamer chambers, the bubble chamber, and so forth. For reasons which sound strange today, the budget did not create competition among the groups.

Lynch:

One of the really interesting things in this era was, this was the beginning for commercial electronics as opposed to building your own lab in your own garage stuff. How did that interact with the users? Was Sugarman and company planned as part of the experiments then or did that phase in as the experiments were built?

Panofsky:

I don't think there was an explicit decision because again, things were quite decentralized. The electronic detectors were provided by the MIT group, Taylor and his people did the instrumentation of the control room for End Station A, and it was probably the earliest time in which commercial computers were used for data analysis. I mean, that was at that time perceived to be a major innovation. Taylor in his lectures likes to point out that the total analyzing power of those commercial computers which he introduced were basically analogous to what we now have in our washing machine in terms of actual computing power, but nevertheless it was a major departure and innovation at that time.

Lynch:

What other proposals were there that were not part of the finished project. We talked about 3 spectrometers. Were there other proposals at the time?

Panofsky:

Not for doing elastic or inelastic electron scattering. All the other proposals had to do with secondary beams. At that time a whole bunch of calculations were made for secondary beams, and the secondary beams intensities were very much larger than we had anticipated due to the Drell-Yan process. I did calculations on neutrino fluxes and even they looked somewhat interesting although as it turned out we never used them. We wrote some technical notes on that. The main calculation on the secondary fluxes were such that it became very clear that, not in terms of total yield of secondaries, but in terms of flux per unit solid angle in the forward direction, that SLAC was quite comparable to secondary beams from the then existing proton machines.

We got independent proposals to do various things. We got independent proposals from Mel Schwartz to do the charge asymmetry of muons, we got a proposal from Mozley and his group to build a streamer chamber which was then very innovative. Then the proposal jointly by Ballam and Luis Alvarez from across the Bay for transferring the bubble chamber was very attractive to me because the high repetition rate. Since the fluxes were entirely adequate and the bubble chamber was not flux limited anyway; therefore the higher repetition rate simply meant higher data rate in a very direct way, relative to the Bevatron which had a few pulses per minute, our rate was so high that the rate was simply controlled by the expansion rate of the bubble chamber rather than by anything else. We received proposals but that was definitely not an initiative taken by the lab, or by me, or by anybody else; that was externally generated.

Lynch:

Let's get back to the End Station A experiments. The Deep Inelastic Electron Scattering experiments eventually resulted in a Nobel Prize for Taylor, Friedman, Kendall. To what extent was that experiment, the result of the experiment, something which was planned or was it just a happy result of the experiment?

Panofsky:

It was not planned as such. The fact that elastic and also inelastic scattering would be interesting, that was planned. At the High Energy Physics Lab, of course, Hofstadter got the Nobel Prize on elastic scattering; that was definitely a planned experiment. What is not terribly well known, there was an experiment done at the High Energy Physics Lab by Ernie Allton and myself on inelastic scattering at the first 3-3 resonance, measuring directly the form factors of the first excited state. So the fact that inelastic scattering existed and was interesting was known but there was certainly zero planning to the best of my knowledge for extensive mapping of inelastic scattering or even a pre-speculation that inelastic cross sections at high momentum transfers would turn out to be as large as they turned out to be.

Lynch:

So people did not speculate "maybe there are seeds in the strawberry jam, let's go look for them?"

Panofsky:

Not to my knowledge. Again my recollection may be wrong on that but I think this was a case of a quite clean, firstly purely experimental revelation, simply that the cross sections held up at large momentum transfers was something which I don't believe was prognosticated, though I may be wrong about that, I may be unfair to the prognosticators. To the best of my knowledge the local theorists did not prognosticate that and in the early decisions for planning SLAC it was certainly not in the dialogue. It was all based on speculating on how the elastic form factors would behave at high momentum transfers.

Lynch:

Would inelastic just be a natural extension of the elastic?

Panofsky:

There had been some smell on inelastic before, so the fact that one would look at inelastic, was in the picture, but it was not in the picture that that would be an unusually fertile ground for discovery. I think that was not in the picture. But again, this recollection certainly has a fair chance of being wrong.

Lynch:

What involvement did you personally have in the End Station A experiments and when did you back away from the experiment to become a full-time director?

Panofsky:

I was personally much involved in the optics of the magnets and I was personally involved in the elastic scattering data taking and analysis, and at that point I gave up because I felt it was unfair to younger and hard working characters who were putting their whole lives into it, so I basically backed out. This was a reasonably conscious decision.

Lynch:

Was it hard?

Panofsky:

No. Sort of natural.

Lynch:

Given all you have put into it, I could imagine it might have been hard to back down.

Panofsky:

Well, I participated in the first experiment. As I said before I was involved in the fundamental optical design, I was not involved in counters, in the detection devices other than in discussions, but not in any detailed design. By that time the drift chambers were invented, and so forth. That was sort of the beginning of a new era.

Lynch:

So, these were striking new results that came out of deep inelastic scattering. Such striking results are the dream of every experimenter, but the more striking they are, the more exciting, but at the same time the more striking they are the greater the burden on the experimenters to make sure they haven't made a mistake. What was the atmosphere in the laboratory? Had the people heard of these results?

Panofsky:

I don't think anybody was worried much about making a mistake, because these are not the kind of discoveries where you are statistically marginal. There was a huge block of data which simply was orders of magnitude higher than you would expect normally at large momentum transfer, so there was excitement but then of course there was a real, to some extent, partition of labor. The experimentalists were excited but certainly most of the interpretation effort basically concentrated on the third floor people in the Theory Division. Of course there is the well known episode of the interaction between Bjorken and Feynman. Bjorken recognized that there was a point-like structure and scaling and all that, and Feynman did the more detailed calculation on partons and parton distributions, and so forth. I had the feeling, I may be wrong, that the analytical engagement by the experimentalists themselves was relatively weak. That may be an observation which is wrong but certainly the detailed excitement, if you wish, which really filled the third floor of the laboratory day and night almost, trying to figure out what these things meant, was really sort of preempted by the theorists rather than the people taking the data. On the other hand, the experimentalists shared the excitement, they certainly were conscious of their responsibility, but on the other hand there really was I would say almost zero concern that the data were wrong. Because they were quite clean, they were certainly background free, they were easy to repeat, the different spectrometers overlapped in their kinematic ranges.

Lynch:

Wasn't there a concern of their radiative corrections?

Panofsky:

There was a concern of radiative corrections but everybody agreed that it was important to get them right but there was no way in which the entire effect could be radiative correction. That was never in the picture, as far as I remember. Then radiative corrections were calculated in detail and that caused a certain amount of conflict; several people calculated them wrong and several people calculated them right. The sort of Bible of radiative corrections was done by Tsai and Mo. But I don't recall, you may recall it differently, that anybody ever believed that radiative corrections were responsible for the entire anomaly. It was simply too large in excess at the high Q squared. I do remember there was a fair amount of controversy on how to calculate radiative corrections right. Radiative corrections is one thing where there has to be major interaction between the experimentalists and the theorists because the meaning as to how resolution enters into the calculation of radiative correction is rather sensitive, namely whether you mean by resolution that of the total array or resolution of any one channel. There are lots of tricks where it is quite easy for a theorist to calculate it wrong if he does not understand the experimental set-up. I remember there was a fair amount of flap on that and I remember a fair number of wrong calculations of radiative corrections kicking around.

Lynch:

I remember that very well myself.

Panofsky:

Did you make wrong calculations?

Lynch:

No, I told Wilson at CEA that he'd made a mistake. It turns out he did.

Panofsky:

The CEA scattering experiments were unfortunate because there were troubles in radiative corrections but also there were just mechanical troubles with the spectrometers.

Lynch:

How would compare the atmosphere of discovery then and now?

Panofsky:

Well, again at that time it was slightly more light-hearted in a certain way. During the earlier time I was very much impressed by the fact that very rarely had accelerators been built for the right reason. When a specific goal was identified in the proposal, (that's true of the Bevatron, it's true of the Cosmotron, it's true of SLAC, it's true of most machines), indeed the new machine did what it promised to do, but it's main impact on physics was in something else. The main concrete argument at SLAC was made for the extension of elastic scattering, which turned out to be relatively routine and not very exciting, while the inelastic scattering and then later the addition of storage rings and so forth really showed the real power of the place. So, some people now are much more compulsive in arguing very hard: Will the next energy region really be sufficient to reach the Higgs or not reach the Higgs? That gets to be a matter of life or death, whether the next step will only be a desert without anything of great interest or whether it will it will lead to new discovery. At that time I certainly was perfectly convinced that the discoveries which would happen and the ones which were predicted would be different because that is what happened during that era with essentially all machines. The Cosmotron was designed for multiple pion production and it made its main impact on the early work on strange particles.

So the fact that expectations and reality would be quite different was an accepted fact, and that in turn made you relax and just keep exploring a new range of phenomena without being as much goal-driven as one is today. The initiative for building the different instruments, the instrumental advancement into getting larger reach into kinematic regions was a much stronger motivation than pinpointing specific physical phenomena. That of course made programmatic decisions in the early days rather difficult because we went through the motions in calling new initiatives "experiments" but a lot of them were not "experiments" in the traditional sense of the word, but were basically extensions of capabilities. We use the same official mechanism of approving extension of capabilities as we would for approving specific runs for specific experiments. In a certain way I always thought even though we played the game that it was somewhat illogical that we use the same mechanism in approving say the third spectrometer as we did to approve the next few hundred hours running time on the spectrometer.

We of course set up at that time a mechanism which in modern terms would be called "transparency," namely having the decision-making of either approving a major instrumental initiative or of running time be open. A proposal would be made, analysis of impact of the proposal on facilities would be made, and then the scientific merit of the proposal and the technical feasibility, and the resource impact would all be thrown into one pot and be considered by a Program Advisory Committee which would then officially advise the director and the director would then officially accept or reject the recommendation. We invented that dynamic in order to forestall criticism of having decisions made in a smoke-filled room between lobbyists and the director. But, this was done all in a fairly good spirit. The main difference was that the resource constraints in the early days of SLAC was during the time when general support of science was increasing about 15 percent per year and so resource constraints were not that hard. The question generally was of approving these new instruments against some absolute concept of their power rather than in terms of a highly competitive situation if you approve one thing you have to disapprove something else. So that enormously changed the atmosphere. I don't recall in those early days that I was ever confronted with a decision whether to kill A in order to give birth to B.

Lynch:

That's a big difference from today.

Panofsky:

Indeed.

Lynch:

Today, there are on-going controversies of what is true and what isn't true. Do you see a difference in the attitude of how critically people look at experiments today versus let's say in the 70's.

Panofsky:

I don't know. We were always internally very critical because one theorem which I felt very strongly about was that one thing which is extremely expensive in high energy physics is a wrong experiment, because once you have a wrong experiment then it takes four or five right experiments to restore confidence as to what the right answer is.

Lynch:

The famous split A-2.

Panofsky:

Yes, the famous split A-2 and the wide-angle electron-positron pair production. They were two classical experiments which were wrong and gave a fair amount of excitement to theorists and it took really a major effort to restore credibility. So we were being very critical; we were conscious of it. On the other hand, the nature of the kind of the early experiments at SLAC were such that one didn't really deal with new results with marginal statistics. So, in order to get the wrong answer you weren't worried about whether you reinterpreted the meaning of a likelihood function correctly and all that, but whether there was something gross the matter. And the thing which you mentioned about radiative corrections is an example. So, we really weren't terribly worried about that. I remember the experiment of Mel Schwartz's on the decay asymmetries. We were really very pleased with that experiment. At that time what it revealed was new.

The asymmetry was predicted but as it statistically accumulated it showed the power of the really high intensity of the secondary fluxes which we had and it was a nice experiment. Had it been done earlier it would also have been a great discovery but it wasn't, but it was a very nice experiment. It was certainly totally unplanned and unanticipated in the original proposals. The next wave were the bubble chambers. The rapid cycling chamber, which was designed by Ballam and Bob Watt and company, was really a major achievement in terms of repetition rate of a chamber. Then Luis Alvarez converted his chamber to a relatively high rate, a few per second. So I had to approve the transfer of that chamber across the Bay and Alvarez had to convince his lab director of that. Again that was quite non-controversial because the particle fluxes by that time were known, the repetition rate of the bubble chamber had been established in the conversion of the 72 inch into the 82 inch chamber so we became for a while really the most powerful factory of bubble chamber pictures in the world. At our peak rate we would produce something like six million pictures a year so we would saturate analysis capacity in many parts of the world.

We were happy about that because SLAC had the reputation of not being a truly user facility because our work was centered around the initial pieces of equipment and, since the in-house people did most of the work on designing and building the major experiments, we were always accused with some merit that the fraction of the work which was being down by outside users was lower at SLAC than at some of the other labs. But when the bubble chambers arrived, then depending on how you would play with the statistics, that would dramatically change that ratio because we would basically saturate all the bubble chamber analysis groups across the country and even to some extent in Japan.

Lynch:

Originally SLAC was conceived as an electron machine and the proton machine advocates did not have a particularly high regard for that, is that correct?

Panofsky:

I wouldn't say it was high regard, it was really an almost division in attitudes. I never understood this very clearly. We inherited various tensions; part of the tensions were actually were not electron and proton, some of them were east-west tensions which dated back to the tensions between Lawrence and Rabi, who had very different views on how physics should be done and all that, and the establishment of Brookhaven when Berkeley was going strong. Also, historically it is true that high energy electron machines had really not made major contributions to particle physics before SLAC. There are some exceptions to that, of course; the pi zero was discovered on the electron-synchrotron in Berkeley. The early bevatrons did photo nuclear reactions but the pion discoveries turned out not to be correct. When SLAC was started there was major skepticism in the proton community. But again, because of the fact that resource competition was not such a large business, there was more of an attitude of, well if the fools want to do that let them do that. It was more tolerance rather than opposition. It was acquiescence.

So the major discoveries at SLAC were certainly a huge surprise to the community. I must say it was also a surprise to the inmates at SLAC to a fairly large extent. Once this happened then there were general arguments which were often voiced at international meetings, whether electrons are really the best tools because they are exploring the unknown with known forces rather than unknown with unknown forces, and all that. But they were sort of ex-post facto rationalization of what really happened. It is still going on today. For instance the fact that even secondary Hadron fluxes became competitive at SLAC really blurred the argument tremendously. Before that happened we used to agree ourselves that classical Hadron physics would remain the province of the proton machines, and that turned out not to be true. Quite a few of the bubble chamber experiments, for instance, made very substantial contributions to classical Hadron physics, although not to the major discoveries in strange particles and so forth.

Lynch:

If there was a viewpoint that, let these guys do what they want to do the conclusion is that it's lower class physics or lower expectation physics, at what point did people accept the fact that electron machines were real laboratories as opposed to second class.

Panofsky:

It depends on the people; some people never did agree that electron machines were major contenders in high energy physics. I don't know. I don't think that has a general answer. I think there was certainly a shift in view in that respect. Of course the whole fact that a laboratory like CERN made LEP a major step, that's partially due to Burt Richter's persuasion directly at CERN. Here was a major laboratory which made its reputation in Hadrons which decided to make an electron-positron collider to be its next major step. I think at this point the fact that you've got to go both ways I think is generally accepted, although I have no statistics of that transition.

Lynch:

You don't see any particular thing that there was some discovery that said, okay, these guys are now part of the club.

Panofsky:

I don't recall anything particularly discontinuous happening there. You know there were the annual meetings on high energy physics, the Rochester based conferences, and certainly they kept having sections on electron and photon physics. In fact one of the rather amusing things is that, as you know, under the International Union of Pure and Applied Physics there are now two series of international conferences. The conferences on high energy physics, which are annual, and the conferences on electron - photon and weak interaction physics, which are held every two years. The thing which became painfully evident in organizing both of these was that it became extremely difficult (I became involved in some of those) to find different agenda for the two conferences. The agenda of those two series of nominally different topical conferences became essentially degenerate, became essentially equal. That is still the case today; they mainly differ more in instrumentation and tools rather than in interest in physics. So, I think most fair minded people would certainly agree that right now things are simply complementary in terms of tools but congruent in terms of goals.

Lynch:

There were some very speculative experiments which were done looking for long-shots. Examples which come to mind are Schwartz's "black hole," the axion search, and so on. How did you decide how much time and resources to put into them?

Panofsky:

I had sort of a principle, which was not enforceable, that one should spend something like 20 or 25 percent of experiments on purely speculative things. There was the axion search, the "black hole," there was Marty Perl's early fairly extensive hunt for new leptons. Of course we all realized that for instance those lepton hunts required fairly large resources but would actually only cover a relatively narrow band of discovery because one could look for long-lived leptons only. It was essentially a mass spectrometer, but in order for it to measure new leptons they would only have to be long-lived to get through the spectrometer. The probability of having them be new and not very rapidly decaying into something else at the same time, was very low. So they were really relatively low probability searches. The axion search was not of that low probability, but again intensities were fairly marginal for that. That search and the black hole were relatively low resource experiments. They were basically almost private experiments done by the individuals, gave negative results and gave new limits. They were not useless at all. I quite deliberately wanted to encourage those to some percentage so that we would not become really a straight programmatic lab where we just programmed running time around the major detectors. We did a fair fraction of that; for bubble chamber allocation you just decided what parameter beams to use. Then of course there was some real innovation there. For instance in the bubble chambers there was the back-scattered laser beam which was proposed by Ballam and collaborators. In the 82 inch it was by and large a matter of simply choosing which particle and what fluxes were in the beam; the process was semi-programmatic. I didn't like to have a hundred percent semi-programmatic laboratory and encouraged various far-out experiments. I had something in my mind like a 20-25 percent concept. There was never a policy; it was all proposal driven.

Lynch:

Did you have an excess of such proposals where you had to deal with them.

Panofsky:

No, actually not. The people themselves, since they were investing their own time, were being quite responsible, so the answer is no, I don't recall ever having turned back far-out proposals. There may have been some monopole search experiments but I just don't remember. I don't remember turning back such experiments.

Lynch:

There was a major change in the way the laboratory worked when colliding beams came on the scene in the mid-1970's. How and when did you first become aware of the advantage and the feasibility of colliding beam experiments?

Panofsky:

That pre-dates SLAC. There was the electron-electron ring at HEPL and Burt Richter was a partner in that. There was Barber, Richter, Gittleman and O'Neill. That did get results and Burt had that in mind. In fact the use of electron-positron collisions was already under discussion in the early days at HEPL. O'Neill talked about that quite a bit. So, that was really not new as far as ideas are concerned; it was only a question when there was initiative to really do it, but that certainly largely due to Richter's drive. Of course we went through that episode. Richter wanted to build it; he made a proposal - not very detailed - but by that time the original money for additional equipment at SLAC had been spent, so this was a new initiative and I tried to get more money for it.

Lynch:

Are you talking about SPEAR now or a predecessor?

Panofsky:

SPEAR. As far as I remember, I may be wrong about this. There was discussion. The electron-electron machine at HEPL was funded, it worked, it did set new short-range limits on a limiting parameter in the electromagnetic force, so it worked. Everybody was aware at that time that electron-positron colliders were powerful but I don't remember that there ever was a serious initiative to build an electron-positron machine at HEPL or in the earlier days at SLAC until Burt specifically pushed SPEAR. I may be wrong about that but I am not aware of any earlier initiative, are you?

Lynch:

I have a recollection when I was a graduate student of Dave Ritson talking about a storage ring, and this predated SPEAR by quite some time, specifically at SLAC.

Panofsky:

Yes, talking about electron-positron storage rings - that I know O'Neill did. O'Neill even talked about one at HEPL. So, talking about electron-positron machines was clearly in the picture. But, a real concrete proposal, including the proposal that the proponent would put a large part of his professional career into the business, I don't think predated Burt's initiative, although I may be unfair to others. I don't remember ever having been lobbied by anybody, including Dave Ritson, specifically to build one at SLAC before Burt did.

Lynch:

Were there not also some other attempts before the final SPEAR that came into being, that Burt had tried a couple of times and was about to give up and Matt Sands said try once more. Was that the one that succeeded?

Panofsky:

Yes Burt tried several times and I tried to help and I flatly failed getting it supported. Then basically Burt and I conspired to do it on the "cheap" by building it without a housing and only shielded with blocks and placing it simply on the tarmac at the end station. So we just managed it on a "pay as you go" basis. I remember having the famous encounter with the controller of DOE, Mr. Abadessa.

Lynch:

AEC or ERDA, whoever it was then.

Panofsky:

AEC - Mr. Abadessa was his name. I said I wish to announce the discovery of an unauthorized particle on an unauthorized storage ring, because we never did actually get a formal authorization for SPEAR. Certainly Burt deserves credit and he led his team to build SPEAR. After the successive failures of getting it supported, we agreed on the "pay-as-you-go scheme.”

Lynch:

So this was basically handled "under-the-table" without AEC's...

Panofsky:

No it really wasn't "under-the-table" in the sense of secrecy, it was with acquiescence of Wallenmeyer and Abadessa as far as that goes.

Lynch:

But not officially authorized?

Panofsky:

It was never officially authorized; on the other hand it was also not in any way clandestine. There was no way to do that. It was not forbidden and was done with acquiescence by the cognizant government people. Again, we did many things with acquiescence. There was no opposition.

Lynch:

How did you choose the people who played key roles in building SPEAR? For example, Burt, John Rees, and so on?

Panofsky:

Well, Rees was, of course, essentially Burt's left or right hand or whatever you call it. Burt really collected his people with whom he was acquainted, whom he knew very well. So, this caused a minor problem to me as director in the sense that some sort of split in the machine-accelerator business between the circular culture and the linear culture arose. There was relatively little overlap between the people who were hold-overs from the linear accelerator design and the people who designed the storage ring. The people in charge of the RF system of SPEAR had essentially zero interaction with the microwave RF people from the linac. SPEAR was really probably the most cost-effective machine ever built in terms of ratio of physics output to investment, but it was done basically by Burt collecting people he felt comfortable with. I was perfectly happy about that but I had some difficulty with, I wouldn't call it friction, but a certain amount of tension between the circular people and the linear people.

For technical reasons we had to unify control rooms which were initially separate, and we had to unify safety systems and beam containment systems, and all that kind of thing. So somehow or other, getting a real integration between the machine operating people working on SPEAR and the LINAC was never quite done. There are vestiges of that even now, so getting a real unity of culture between those two proved a little difficult. So SPEAR was really pretty much a near autonomous enterprise on Burt's part and he did it extremely well and he had people loyal to him. We did not try to do matrix management, namely trying to have the work there done by assignment from the central groups, but it was pretty much a closed group. That was also true, incidentally, of the bubble chamber group. They had by in large their own technicians, their own engineers, or at least on very long-term loan. We never did do very much what today is called matrix management namely assigning professionals in electrical engineering, electronic engineering, mechanical engineering, software development, and so forth, from central groups with divided responsibility between the heads of the central groups and the heads of the instrumental teams. We reorganized at various times to simply reassign or transfer people.

Lynch:

In the choice of the parameters of SPEAR, was that more or less freewheeling or was it set by economics or physics goals?

Panofsky:

Not by physics goals. That was one of the interesting things considering the history of discovery. It was set basically by geometry and economics and then, as you know the scaling laws vary fairly steeply with energy. It was set by the fact that we wanted to use off-the-shelf RF tubes. The thing had to geometrically fit into the target area without any civil works. Burt would probably agree with this observation. It was not focused on specific discovery reach.

Lynch:

Just whatever you could get in the real estate?

Panofsky:

It was set basically by economics and geometry.

Lynch:

SPEAR had two interaction regions for experiments, meaning that two experiments could run simultaneously. How did you choose which experiments to allow to be built and to run?

Panofsky:

I made a very deliberate decision to have one interaction to be a general purpose detector and to have the other interaction region be used for successive dedicated experiments. In retrospect it is not clear that that was a good decision because several dedicated experiments were done in the second interaction region which in some cases were actually less powerful than what the general purpose detector measured in the other interaction region. As an example, there were two dedicated experiments: Hofstadter had a dedicated experiment with sodium iodide detectors to measure to e⁺e^- into gamma-gamma rates and then O'Neill had an experiment where he had a relatively small angle spectrometer arrangement to measure e⁺e^- into Kaon pairs and pi pairs, but where the reach in momentum was higher than the reach in momentum which was possible in the general purpose, the so-called MARK I, detector which Burt Richter and his collaboration built. I felt it was a bad idea to have both interaction regions be used by general purpose detectors and basically made this a ground rule for proposals.

As I said before, it may have been a bad decision. On the other hand having two very similar detectors would not have added very much to knowledge either. So, the dedicated region produced data, but really only very minor increments to what was done in the MARK I. Certainly the main impact on the physics was done in the MARK I detector. There was one experiment in the other interaction region which was an outright failure for which a proposal was made in weak interaction physics. I won't go in detail. By and large, the dedicated more single function experiments, turned out to be not very productive. For instance, in the O'Neill experiment, simply because of intensity, the total number of events which happened at energies higher than the ones accessible to the MARK I, but which were accessible to the O'Neill machine, were very small. In the Hofstadter experiment, even though the gamma ray conversion efficiency was much higher, in his detectors the solid angle was correspondingly smaller; that also didn't add very much.

Lynch:

There were very few dedicated experiments then and after. There was a dedicated experiment at DESY - DASP for example, but that was basically the end of dedicated experiments. Are you saying that basically the general purpose detectors took over the world just because they had the solid angle?

Panofsky:

Yes, I struggled against having the general purpose detector taking over the world, but failed. We learned that one must be very critical about non-general purpose detectors that the special thing which they chart out really competes with the general purpose detectors and at SPEAR they just didn't. It was a mixture of things, but fundamentally it didn't because of limited solid angles and the rest of it. There was one case which was just not a very good experiment. So indeed, the MARK I basically dominated the physics, but then once it did then you have the question, how do you do independent work on the general purpose detector? So, the two groups — Burt Richter's and Marty Perl's — essentially combined forces and you might call it general purpose or not general purpose, but Marty Perl proposed to add an external wall for gamma ray conversion externally to the MARK I. That was quite successful and became integrated into the MARK I system.

Lynch:

Now the initial experiments were interesting, the measurement to the total cross-section and so on, but the world of high energy physics became very excited in November 1974 with the discovery of a particle called the psi at SLAC and the J at Brookhaven. How do you see this changed the direction of high energy physics and the attitudes of high energy physics.

Panofsky:

Firstly, things became already "exciting" even before one went up to the J-Psi situation when the total Hadron cross section didn't really behave right, the way people expected it should. It showed already some rise in the ratio R between the hadron products in cross-section and the purely electrodynamic cross-section before it went to the J-Psi and we didn't quite understand that. There was sort of an atmosphere that things weren't quite right the way they ought to be, for quite a while. Then of course came the actual discovery itself and I think that's been described and documented ad nauseam. The famous fact that because Schwitters was a lousy operator he didn't exactly follow the energy scan which he was supposed to, and all that, and I was gotten out of bed and we all gawked at this. The main thing which it demonstrated to me was something which I considered to be very fundamental is this.

All discoveries involve a new phenomenon and the credibility of that phenomenon increases in time. In the experiment at Brookhaven the credibility increased very slowly in time, simply because statistics accumulated gradually and one had to understand backgrounds and so forth. In the J-Psi discovery here at SLAC, the credibility went essentially from zero to infinity in two days. So then, when you look at priority of discoveries, the question then comes what is your cutoff, what is your threshold of credibility at which you call something a discovery. Clearly if the Brookhaven group had the courage or foolishness to believe the earlier data, they could have had much more priority of discovery. On the other hand, since they were in fact being careful and kept it very secret, the priority didn't come out that way. So very often priority disputes have their root into where you define the threshold of credibility. If you define the threshold of credibility low enough you sometimes find something earlier but you are not sure of it. There is some arbitrariness to this, so it gave me a fair amount of insight in this. Of course it became quite controversial because of this fact, you know the history of this.

Lynch:

Let's get to that one later.

Panofsky:

Anyway, as far as the whole business of priorities is concerned, clearly it was sort of a classical example of priority versus credibility. I was at CERN actually when I heard about the earlier data, where there seemed to be a rise in the ratio R. And in fact I gave a seminar on it and there were lots of theoretical speculation, before the J-Psi was discovered. In the early days at the Italian storage ring things were fairly mixed up but the hadron cross-section also appeared to be uncomfortably large. So, not only was the question of discovering the j-psi, there was also the question of what are you discovering, namely are you just discovering enormously large total Hadron cross-section or are you discovering a specific new state. It gave a fair amount of insight that when you argue priority and methodology and all that, that to some extend there are lots of definitional problems about what really do you mean by this discovery.

Lynch:

Those things can certainly be very complicated and people's egos get mixed up. What role did you play in deciding when to go public?

Panofsky:

In this particular case, there was zero problem, the fact that there was a peak was obvious in one day; people began writing their paper during the data taking process. There wasn't any decision to be made as far as publishing is concerned. There were a lot of decisions to be made about publishing simultaneously with the MIT group and their announcement versus publishing because it was very spectacular. Fundamentally one did not expect narrow states like that theoretically in advance.

Lynch:

You've mentioned the fact that Sam Ting and company at Brookhaven discovered the same particle at the same time. How did you learn about this, and what was your reaction?

Panofsky:

From Sam Ting. He was a member of our program committee and he happened to be at a meeting here of the program committee at the same time when the peak was discovered here. I remember hauling him into my office and Burt's people gave him a lecture on the new data and he visibly paled and he then told us about his experiments. We then decided to have a joint or parallel, I don't remember which it was, seminar here where Ting discussed his stuff and Burt's group discussed our stuff. It was a singular event because as I said before, credibility of something major happening shot from essentially zero to a hundred percent in one day. So, the MIT group had a fair semblance of a peak in the e⁺e^- pairs from Hadron bombardments of beryllium for quite a long time, for several months, but they were just arduously accumulating statistics and running checks and Ting runs a very tight shop.

So they kept it extremely quiet because Ting basically got his reputation by having shown that Pipkin's experiments on wide angle pairs were wrong and the very last thing he wanted to do is come out with a wrong experiment. He had very high standards in terms of really wanting to believe it was right. But there is a direct tension between having high standards between believing it is right and being early. When he heard about this he immediately burned up the telephone to his home group and people went public. There were several really strange situations. The fact that both groups did the work independently is clear; the fact is that in one case credibility rose slowly and in the other case it rose rapidly. The fact, however, that Ting happened to be here as a member of the program advisory committee on the day when this happened was just chance and sort of fun. What else can one say? It was one of those wonderful days of physics but also clearly it hurt Ting to some extent because he had been laboring for a long time on this particular thing and then suddenly it jumped out in front of his face, being done here. So clearly he was not a happy man.

Lynch:

I believe that. One of the eventual outcomes is that both Burt and Sam got a Nobel Prize for the discovery but one of the sad side effects of the discovery was a rather acrimonious atmosphere which developed between the two. You hinted at this earlier; would you put that in context with the present discussion?

Panofsky:

Yes. What happened was that a rumor spread in the East that in fact the tentative presence of this peak, which had not been publicly announced, had leaked here and that therefore the people here had set the energy at the peak and therefore scooped, if you wish, the MIT group. That rumor was so intense that some of the senior people at MIT actually believed it, including Victor Weisskopf and Martin Deutsch. I had visits with both of them and I can certainly personally certify that the rumor was not true. I can certainly certify the fact that the energy settings were more or less fool's luck, that in fact there was a scanning scheme of regular intervals in energy that may have missed the peak had it not been for the fact that they were being inaccurately set. I was personal witness to that, but there certainly is no question that a vestige of the peak was present at lower credibility before. It was Ting's own choice to keep it secret until he really firmly believed it. I give him credit for that; that's a sign of high standards. But the reason for the acrimony is that this rumor was spread in the East and I got personally involved from personal witnessing, but not participating, in the measurements here that this simply was not true. I had a session with the two eminent gentlemen from MIT who were somewhat perturbed by all this.

Lynch:

One of the things I remember from that time was an article which you wrote, I believe, for "Science" which I described as some famous lines from Beethoven's 9th Symphony, "O Freunde, nicht diese Töne. Sondern lasst uns angenehmere anstimmen und freudenvollere." That's the way I …

Panofsky:

I don't know what article you are talking about.

Lynch:

This was basically in response to an article which Sam had written I think in "Science" which was describing the controversy and you wrote a response to that, saying "Look folks, let's rejoice in the physics that we found here instead of fighting over it."

Panofsky:

That's right. I didn't want to do then in the publication what I was saying here. I didn't want to give a blow-by-blow description of exactly negating things and so forth. But I really believe what happened here is something entirely normal and reasonable, namely that because that's the way it is. The MIT experiment had a very low counting rate and therefore it took months to accumulate enough data and statistics and background checks. There was a problem in Ting's instrument about pi versus electron rejection. All these things had to be really right because he had a very high demand for particle identification. So, he worked very hard to believe his answers and we were just damn lucky. There is absolutely no question about it. The discovery here was fortuitous, but fortuitous is probably the wrong word. We did an energy scan and the energy scan showed it up.

Lynch:

But it could easily have been missed.

Panofsky:

It could possibly have been missed. Probably even at the coarse energy scan if it had been done very precisely the radiative tail of the peak would have given high readings but not a sharp peak. So the question is: had that happened, would people then have decided to make more narrow scans? It is a "what if" question. I don't have any idea; maybe you can get that straight from Burt. I don't think it matters because that isn't the way it happened. The way it happened the scan was carried out somewhat imprecisely. But there is absolutely no question that it was influenced by a security leak from Sam. His security was very good. And again, to some extent I found nothing blameworthy on either party and one should rejoice. But the nature of the two experiments in terms of data rates were just so vastly different that it was simply a fact of life that one had a slower accumulation of confidence and the other one the confidence just went "bang." That of course certifies to the power of electron-positron storage rings. Doing electron-positron interaction experiments in the final state by producing them first with Hadrons is a much harder way than starting the initial state from lepton-anti-lepton pairs. And that certainly was demonstrated by these events so that's not a question of individuals. It has to do with the physics.

Lynch:

Do you think that that fact changed the direction of high energy physics?

Panofsky:

Oh, I think so. I don't know whether people logically look at this one experiment, but certainly all the data explosion which then followed in terms of the whole spectroscopy of the charm anti-charm states did. One couldn't imagine getting that kind of spectroscopy in that short of time with that precision by looking at what amounts to lepton pairs in the final state from initial hadron-hadron collisions. So that whole pattern, I am sure, which then evolved in terms of the whole onslaught on the spectroscopy of charmonium certainly changed the picture of physics.

Lynch:

There were numerous other discoveries that came out of SPEAR, the charmonium states and open charm, which you just mentioned and the tau. We haven't talked about the tau yet. That eventually resulted in the Nobel Prize for Martin Perl. Now this discovery was much more subtle than the psi. How was this discovery received inside at SLAC and outside?

Panofsky:

Here was a case where the opposite was the case relative to the J/psi discovery. The discovery was not sudden but there was basically a small excess of electron-muon coincidences. Perl actually did the opposite thing to Sam Ting. He did not keep the initial data secret but he went around to neighboring labs, including Berkeley, presenting his data. He even presented seminars saying, I don't know what this is, it may be a sequential lepton but it may be a lot of other things. So, he tried very hard to proselyte non-believers in his experiments. He used the very opposite social dynamics of keeping it secret until he really believed it. He initially himself didn't believe it, but he was quite sensitive to it. And of course Paul Tsai had made the exact cross-section calculations, assuming there is a new lepton, what the cross-sections would be, how would it behave near threshold, and all that.

So, at that time it was very different; here was a clear theoretical model for a third lepton, but Perl didn't know if it existed. If it existed you would know what to compare it with, but the initial statistics were marginal. But I thought it socially very interesting because Perl was a collaborator with a whole group building the machinery here, but he did the mining of the tapes, essentially, for this phenomena with a very small group, so that was very interesting. I thought, by the way, that one of the most important discoveries which really hasn't been heralded as much, was the determination of the spin of the quark by the discovery of conversion of the e⁺e^- into two jets and showing that the angular distribution of the jets corresponded uniquely to a spin of one-half of the quarks. I think that was one of the really most important experiments next to these two spectacular discoveries. It was an unusually fertile ground. Of course the tau was quite delayed in terms of its revelation, if you wish, just because Perl was basically shopping for other explanations of his data and eventually filtered it down that this must be it. I think it took well over a year to two years to do that. Again, I don't remember exactly.

Lynch:

You just mentioned that, you didn't use the word but I'll put the words in your mouth, SPEAR was extremely successful. Now the question which I have, was SPEAR too successful in the sense did it generate jealousies inside SLAC and outside?

Panofsky:

Well, I don't know. That's hard for me to calibrate. Inside SLAC, not really because everybody was sort of basking in the general glory and in the warm but diffused sunshine or whatever. And of course, SPEAR was not unique. The discovery of the basic quark parton picture had preceded that. I don't think so. I don't think there was jealousy; I alluded to this beforehand: There had been a "culture gap," if you wish, between the collider beam and stationary target folks, and the colliding beam folks and the linear accelerator folks. This gap was more on the instrumental side than on the particle physics side because the techniques are so different. I also alluded to that, that classical techniques for dedicated experiments had failed. So therefore there was a split in cultures, both on the fixed target collider side of things and on the technical side of things. I'm not aware; maybe I'm being naive; I'm not aware of "jealousy." I am aware of questions, for instance, when decisions had to be made, who would become head of accelerator physics or head of operations, and so forth. How to choose among the two cultures; there were some difficulties in deciding the leadership in the accelerator-collider organization here at the lab. Certainly when Dick Neal retired he certainly deserves enormous credit for having built SLAC on schedule, on budget, and all that. He was a fantastic organizer and a very good solid, non-highbrow engineer. But he felt somewhat put upon by the sort of free-wheeling characters of the "circular" type. But he was too much of a gentleman to do this advertently. But when he retired, he basically really retired, which other people did not often do.

Lynch:

We talked the question of jealousies within SLAC. Now there's a possible flip side of that, and that is, given the enormous historical success of SLAC, in hindsight did that create an atmosphere of hubris?

Panofsky:

Well, one can maybe argue, but probably not with much merit, that the whole PEP authorization, construction, building that many interaction regions, building that many detectors, you may describe that by over confidence. It was certainly an over investment considering the total productivity. Here SPEAR was enormously productive with one detector and in an energy region which turned out to be extremely fertile. Then, PEP was proposed; I supported it. Then, by that time, we got proposals for many multipurpose detectors, and in retrospect it was probably over confidence in the discovery potential of PEP that we approved so many detectors simultaneously. In that over confidence, clearly we were joined by our advisory committee, and by the fact that by that time confidence in the productivity of that approach had spread and so many people wanted to get in on the act. In a similar situation one can argue maybe that there are too many detectors at LEP, because the physics that you do with the different detectors is not really that different. So, I think your remark has some merit in respect to PEP. I think PEP was to some extent a product of over confidence. There was no logical reason for that particular choice of energy. One just expected that by extending the kinematic boundary you would again strike gold, and one didn't. We approved several detectors, HRS, TPC, MARK II and DELCO. That was probably an over investment, with the benefit of hindsight, but they all did good physics, but none of it near the general major significance of what happened at SPEAR.

Lynch:

One of the big differences between SPEAR and PEP was that ,while SPEAR was built entirely by SLAC, PEP was a collaborative effort with LBL. How did that come about?

Panofsky:

Well, that was done by negotiation between me and the director of LBL, because LBL by that time, of course, had the Bevatron but it was coming to the end of its lifetime for high energy physics, and they wanted to keep their oar in. That collaboration was not, in my view, extremely productive as far as the machine is concerned because the knowledge of the detailed technology was largely here and paradoxically the main contribution to the construction of PEP by LBL was in civil engineering, which is sort of strange: the off-site people did most of the site engineering, while when it came to hardware, essentially no hardware was built for the basic machine at LBL and brought over here. They contributed some very valuable people and very good people, both in installation and civil engineering, but they did not contribute in a major way to the RF or magnet design or even orbit dynamics. There were relatively few people in the collaboration who learned the details of the orbitry. So, it was not a strong collaboration; it was however, a major contribution.

Now when it came to the detectors, then of course some of the detectors were almost entirely built away from here. The TPC was built essentially by a collaboration headed by LBL and essentially nobody had a hand in it here. The HRS was more or less designed and built at Argonne National Lab, and so forth. So, again it was a social shift from SPEAR very much because (a) there were many detectors, (b) it was a collaboration in construction but still with very heavy SLAC domination. In addition to that, because of its size, we had a lot of trouble in civil engineering. We had a bad contractor and lots of controversies and claims and so forth. Of course SPEAR was built without any civil engineering and it was possible to do that; it would not have been possible to do that in PEP. So all the vicissitudes in a major construction enterprise, including its architect and engineering, civil construction, collaboration, higher costs and all that, affected PEP while in SPEAR it was simply absent.

Lynch:

In parallel with the construction of PEP, DESY was building PETRA. What was the interaction between the two machine designs and the people?

Panofsky:

Oh, friendly. They were very much aware of course of the competitive situation, but there were lots of exchanges. Jentschke and I became good friends during that period.

Lynch:

Jentschke was the director of DESY at that time?

Panofsky:

Yes, Willibald Jentschke. He incidentally spent his sabbatical here at SLAC, later participating in the Prescott experiment on elastic electron scattering on the deuteron, which again was one of the major break-through experiments, which we didn't mention. The SPEAR discoveries, however spectacular they were, did not overshadow the whole work of SLAC and the elastic electron scattering on the deuteron was one example which was really a very good experiment and showed what can be done if you integrate knowledge about the accelerator with knowledge of the detectors. Again, I may be naive. I visited DESY quite often during those days, and many people ran back and forth, and all was reasonably peaceful. I don't think there was any real rivalry and certainly there was nothing like the MIT/SLAC situation as there was in respect to the J-psi, partially because there weren't any discoveries to fight over to quite that extent. Certainly DESY deserves the credit for the multiple jet structure indicative of the gluon, but as far as I can figure out there is more rivalry and competition among the different groups at DESY concerning the priority situation about the multiple jet structure than between PEP and DESY.

Lynch:

PEP was conceived before PETRA but PETRA came on line before PEP. How did that come about?

Panofsky:

Well, I don't remember in detail. Partially we were budget limited. PEP was built within total estimated cost but the annual funding got stretched out during those years and I made an impassioned plea to move some of the money from later years to earlier years and finally succeeded. But we also had trouble with the civil engineering contractor; I called him the "late Mr. Early." The name of the civil engineering firm who did the excavation was Early and Company. They were extremely claim conscious people and tried to convert any delays into claims.

Lynch:

What effect was there, do you think, on the SLAC physics program due to the fact that PETRA turned on earlier and produced whatever physics they did before?

Panofsky:

I am not aware of any in particular. It was sort of checks and balances. I really don't understand this quite. Most of the physics results were reasonably complementary. We didn't get evidence of the 3 jet structures, and we didn't focus on that very much. We were more focusing on various electrodynamics limits and leading particle dynamics.

Lynch:

People also measured the B lifetime.

Panofsky:

And the B lifetime, which was one of the more spectacular things; for reasons which are not totally clear to me, we were first on that. The priorities at DESY turned out to be quite different, and that was not coordinated. So, it was really a very complementary situation, although I must say in terms of total of productivity it was somewhat disappointing. The technical effort was very large. For instance the HRS was a huge effort and involved hauling superconducting coils across the country and building a cryogenic plant for that, because our laboratory had done very little work on superconductivity before. This was pretty much of a technological shift for SLAC. We've done cryogenics with the bubble chamber but not with magnets. So it was a bit of a tour de force technologically with not that much to show for it. On the other hand there were some very interesting results. As you pointed out the B lifetime was possibly the most important one but there were many other good results. Of course the initial rise in the Z-zero peak began to show. So, there was important physics; there were a fair number of publications and so forth.

Lynch:

You mentioned the TPC earlier and I don't want to pick on the TPC, but bear with me for a moment while I develop the subject. There were various technical problems and especially some financial problems. I'd like to ask what they were for background, and then ask the important question, was there some long-term fallout from that result in terms of interactions?

Panofsky:

I was involved in the TPC very personally just because there were so many headaches. The TPC was invented by Dave Nygren, which is really a very excellent contribution and there are still TPC's and they are being used for many other applications, including for heavy ion machines and so forth. But Dave, being a very good scientist, did not manage it personally. It was managed by a large collaboration and the spokesman and general manager was elected by the collaboration. So the financial responsibility, the technical responsibility and the management responsibility were highly diversified, namely that the spokesman came from Johns Hopkins, as I remember, but most of the money through Berkeley, so nobody really had total responsibility. The thing was very diffusely managed and I finally decided we were in trouble and staged a tantrum and went to see the director of LBL and we agreed he had to assign a project manager to the project. It couldn't be run by a democratically managed collaboration, the way it was going. That assignment fixed it although after some delays and some overruns. So it was really an interesting example for me. It was run by laissez faire and didn't work because the basic technical concept was very good, was very powerful, but the individual who invented this and who also made the proposal and persuaded the committee and me to accept it, was not the one who actually executed that large scale project.

Lynch:

Now the corollary to that is, you are talking about the specifics of the TPC, but what I'd like to ask, do you think there's a longer term fallout in terms of the much deeper involvement of DOE in oversight of projects? Did that come out of that or is it parallel?

Panofsky:

I don't think DOE did intervene very much. I think it does mean that the host laboratory has to get itself involved much more heavily than at that time we were. Because, after all, the host laboratory, the way it used to be and to some extent the way it still is, the host laboratory simply can decide whether the detector is ready to move in or not to move in. In principle the host laboratory could take a position of hands off, because this particular collaboration did not involve any members from the host laboratory. The lesson I learned, and I don't know whether everybody learned it, is that simply does not work.

I don't think external DOE oversight helps very much either but you've got to have (a) heavy involvement of the host laboratory, maybe by always having scientists from the host laboratory be a member of the team, because otherwise the communications just don't work well enough; (b) secondly, even though it is a collaboration and even though money comes from many sources, some of them even foreign, there has to be a designated project manager who resides at a laboratory, which does not need necessarily have to be the host laboratory; (c) you cannot manage efficiently by having different groups have responsibility for different components and just have coordinating meetings to make sure that the software is compatible or that the plugs fit at the interfaces or whatever. The collaboration has to agree on who is the central manager. We failed in that initially and we pulled it out by doing that. I certainly learned that that is an absolute necessity. Managing large collaborations on detectors is of course very difficult fundamentally, because the sources of money comes from different government agencies, comes from different institutions, comes even from different countries, and still you want to build one device under central management. It's an interesting social experiment.

I don't think having DOE or one of the other government agencies assume more responsibility will work. DOE ultimately cannot tell the Ministry of Atomic Energy in Russia what to do, so DOE doesn't have any particular unique role to play. There has to be an agreed responsible individual for the whole thing who really can spend full time with the staff and whatever it takes to do the job even though there are independent contributors. We didn't do that very well in PEP; we didn't have to do it in SPEAR because all the effort was in MARK I. Even for MARK II, where we had several in-house groups, it was all in-house. We didn't do it very well in PEP. It worked well for some of the detectors because they were responsibly run from one laboratory. HRS was run out of Argonne; it wasn't the greatest or most wonderful detector but it was well run. TPC was a very ingenious innovative device but it was not well run, initially at least, so it had to be bailed out by a reorganization half way through. It simply means that in any future large detector, the most important thing is that the collaboration agrees on a well-defined line of authority under a general manager, who has authority to say when you aren't ready, you need more resources, you aren't performing, in short, who can behave like a project manager of a large project.

Lynch:

And those interactions are difficult with physicists.

Panofsky:

Indeed, and I am sure will continue to be.

Lynch:

In the remaining time let me move to the next major accelerator project at SLAC which was SLC. Just for background can you describe what it was and what was special about it?

Panofsky:

It was generally recognized, and Burt would make speeches about it, but it was again very common knowledge that the general scaling laws pertaining to circular e⁺e^- machines were well understood to be more or less quadratic, if you match the costs which relate to the RF system with the costs which are basically size dependent. Therefore LEP would probably be the last circular e⁺e^- machine ever built. But because of the success of SPEAR, PETRA, PEP, and so forth, one wanted to go higher with e⁺e^- colliders. So therefore Burt initiated a major study on what at that time was called a single pass collider, the SPC. The linear collider is "single pass," signifying that you collide the beams once and throw them away. A study group worked out various proposals. One of these was the SLC. Of course we recognized that this was not a true linear collider consisting of one linac shooting at another linac but that one would take electrons and positrons and bend them.

Therefore some of the radiative effects in that single bend would have to be very carefully studied in order to make sure that the phase space dilution of the single bend was not going to be too destructive as you go to high energy. In addition, for this single bend, the magnets presumably should be designed to go to a higher energy and therefore one wanted to build them as economically as possible. Of course in the meantime, and we didn't talk about that and I was personally very much involved in that, the whole business of getting the energy of SLAC up by the SLED method to the 50 GeV level, would have been impossible in the circular e⁺e^- mode So we had a situation here that we had the potential of going to e⁺e^- energy of 100 GeV in collision and of going up to 50 GeV eventually on stationary targets. But to make collisions there in a storage ring was impossible at SLAC, but since the power of colliding beams had been demonstrated one wanted to preserve that. Between the machines we talked about and the SLC we undertook two ventures which miscarried, namely (a) to make the whole machine superconducing, and (b) the so-called the RLA the re-circulating linear accelerator which is in concept very similar to what is now at the Jefferson Lab, where you get the higher energy by using the linear accelerator twice. And both of those went quite far in conceptual design. The RLA went to an actual proposal to DOE and was not accepted; the superconducting design study was carried out in fair detail, and I am sorry to say I killed it before it ever went to DOE.

Lynch:

Are you sorry?

Panofsky:

No I am not sorry. In fact you can see here behind me is the proposal for the two-mile super-conducting accelerator and the RLA. There were several attempts carried quite far in terms of concepts and then it was decided for better or worse not to pursue them. The SLC on the other hand was pursued but it took a very large effort analytically and several people contributed a great deal: Matt Sands got involved, and many other people, Ritson played a large role to really make sure that the radiative effect in the bending magnets and the final focus aberration would not disrupt the thing. Interestingly enough the beam-beam disruption was analyzed then, but not quite enough. The beam-beam disruption was analyzed classically in terms of the mutual focusing only. The quantum effects and some of the other effects were not at all understood at that time. But on the other hand, the quantum fluctuations noise driving effects in the bends were analyzed in great detail. The SLC we approved, we got money for it, we upgraded the energy of the linac which was fairly dramatic.

We probably underestimated rather drastically initially the demand on the reliability which the SLC requires and its enormous multiplicity of components. The basic SLAC Linac is fairly forgiving of imperfections. For instance, if one klystron goes out you just cut it out and raise the energy of the rest of them a little bit. In contrast SLC is much more unforgiving of relatively minor fluctuations in magnetic power supplies or failures of individual klystrons and so forth. We underestimated very significantly, and that encompasses all of us, the total amount of commissioning time needed to get around all that.

Lynch:

One of the consequences of that was that the SLC was very slow to produce results, both in time and in magnitude compared to the rival at LEP at CERN. How did this affect the image at SLAC?

Panofsky:

Badly. The struggle in that delay clearly, both internally and externally, gave SLAC a lot of trouble. It caused morale problems internally, it caused big worries about priorities internally, namely just how far to go in terms of putting more resources internally into the SLC. Of course, this was not on my watch. It caused a great soul searching internally whether to give up and go on to the next step with a risk of having basically a very long drought here or not. We happily pulled it through and I think the net result has been very positive both in terms of physics results and demonstrating the potential of linear colliders. But that long drought which was due to our under-estimate and the reliability requirements certainly hurt. I don't know what to say about it.

Lynch:

I understand. What do you think the long-range effect of SLC has on high energy physics?

Panofsky:

I think right now there is fairly general acceptance that after LHC linear colliders must be built or should be built. God knows when, where, or by whom. But I think that general persuasion would not have existed without the SLC. Had there been just paper studies and test accelerators and all that, it would probably be much less convincing. I think SLC, both in terms of its albeit limited but important physics results and by demonstrating the reality of linear colliders is a very important contribution of SLAC.

Lynch:

This is an interview with Wolfgang Kurt Hermann Panofsky done in his office at the Stanford Linear Accelerator Center on 2 May 1997 as part of the AIP oral history program. In the previous interview we discussed the physics done at SLAC while you were director. I would like to discuss the unique style which you had as director. One of the characteristics of you as a director was that you were always very well informed. It was very rare that someone could come into your office and surprise you. Not everybody has the faculty and I am curious how you did you do that? How did you have your fingers on the pulse of the lab?

Panofsky:

I don't have the slightest idea. I was just interested in many things and I did a tremendous amount of running around on the different parts of the lab, and read a lot of literature and so forth, but there was no strategy to that or anything of that kind. It was simply due to the fact that I was on speaking terms with essentially everybody in the lab. I also wanted to make it very clear when I participated in discussions, that it didn't make any difference that I was the director, namely that when I had a certain opinion it was just another guy having an opinion. When a meeting was dedicated to make decisions, that would be specifically set out up front. However when there were just casual meetings or discussions or visits or what have you, people wouldn't assume that a discussion had to be resolved in a given direction with the director making decisions. It was made very clear that the director was just one of the technical types talking either sense or nonsense as the case may be. But there was no organized strategy to do what you say.

Lynch:

Actually I would say that is in fact quite organized because the tone that you set in talking with people very strongly influences the feedback that you get. That was one of the things which really impressed me, was that people could talk with you so freely, it was not the questions of the "Pief up in the clouds."

Panofsky:

I always wanted to make it very sure that one would not make decisions at random, suddenly popping out of some discussion which was not designed to make decisions. It was sort of interesting, different people in the lab had different styles. For instance some people in the lab were extremely worried if discussions didn't stay within the organizational structure. For instance Arnold Eldridge, who did a fantastically good job in directing the manufacture of the accelerator, and was absolutely excellent in doing that and was a more old-fashioned organization person, and he was always very nervous if somebody who worked under him would talk either to Dick Neal or myself without him being present, being afraid that decisions would be made over his head. I tried to make it extremely clear that if I had a technical session on the alignment system or something with one of the engineers who formally reported to him, that we wouldn't make any decisions but we would simply try to solve problems. That worked pretty well.

Lynch:

Another part of the director's job is to resolve conflicts between people, and one of the things which I found quite interesting is that the parties could come to your office, there would be a lot of yelling and screaming, and a decision would be reached at the end. That's not unusual but what is unusual is that even the losing party usually came out of the meeting feeling relatively good. How did you do this?

Panofsky:

I don't have the slightest idea. We did indeed have lots of noisy sessions, indeed usually resolved things, and I assumed since everybody had the opportunity to really state his opinion loudly, vociferously, blow off steam, that he would not feel that he was in any way being throttled but beyond that I don't think there was any system to that madness. In fact, I remember several occasions where we had a meeting with a newly hired engineer and a senior member of the faculty and we tried to resolve something or other and the senior gentleman of the faculty said, you are out of your blah, blah, blah cotton picking mind. The engineer thought that the gentleman would get fired immediately or something like that, but the discussion just went on nicely ahead and somehow or other something got decided and nobody got fired and nobody had a heart attack and so forth. So somehow or other the fact that people could blow off without this being taken personally helps.

Lynch:

Okay, some of these things just happened organically. Management style is something which is highly individual and we just talked about one aspect. But another aspect is how one delegates authority, how did you decide how much authority to delegate to people versus grabbing on to things yourself?

Panofsky:

Well again not terribly systematically. Dick Neal was really a tower of strength during the formative period of the project. He had delegated authority to balance - to rob Peter to pay Paul - that means so to speak; that means to shift resources around within the construction project, but at the same time we had sort of a gentleman's understanding that if things like that were bigger than a certain amount, I would be involved. We had a tacit understanding that although Neal had total delegation on the accelerator and the Research Director had authority on the major target area instruments, that there was an unwritten threshold for them to involve me if the decisions involved more fundamental things. That of worked out in practice because as a practical matter I was involved in the technical discussions preceding the decision anyway, so I knew generally what was going on even if the decision was made at the lower level. That really is the main point, that I knew what was going on so decisions were made at the lower level but I was not in any way blindsided by these decisions being made.

Lynch:

So you did not have any formal signature authority for so much money…

Panofsky:

No I didn't have any formal threshold. That means that if something involves more than $10, 000 or $100,000 or whatever, I should be involved. We never had any such formal kind of agreement within the laboratory. It was more or less a matter of judgment either on the Associate Director's part or somebody else's part, or my own part, when the matter clearly was a crucial issue for the laboratory, but there was never anything written down to the best of my knowledge.

Lynch:

That's a different style than the world lives by today.

Panofsky:

The more modern system always tries to formulate exactly the threshold of delegation , particularly between government and non-government. There was nothing of the kind.

Lynch:

In order to successfully delegate authority it's necessary to choose people whom you trust to do the job well. I think everybody would agree that you did very well at that. How did you choose people in positions of authority?

Panofsky:

I don't have the slightest idea. Basically by acquiring personal confidence. Again, I mentioned Dick Neal. He had been around for quite a long time in the microwave activities at Hansen lab and I just became accustomed to him even though we had extremely different styles in personalities and approaches. I basically acquired confidence that he was the gentleman who would do things right and straightforward and make choices in a very deliberate manner. Ballam was a similar situation. I've known him for quite a while in Berkeley and we kept in touch. It was basically a matter of knowing the particular individual for a protracted period of time and then twisting his arm to join us. Now Pindar, in terms of leadership of the business services, had a long and very distinguished history as an administrator at Stanford.

It is sort of interesting - he was originally recruited by Schiff, because in the old days in the Microwave Laboratory excellent technical work was done but all sorts of expenses were incurred which then turned out to be unallowable because they were in violation of something or other, and Pindar had been recruited by Schiff to straighten all that out, which he very successfully did. This was all before my time. It was only logical for him to continue in this. On the administrative side, Moulton, had worked very successfully with Ginzton before SLAC was authorized. The important thing is that we all reached agreement, the senior people joining, that the purpose of the place was to do physics, and therefore we even coined the names - we didn't have a business division but a Business Services Division just to signal that law and order was not an object on its own right but that it was to serve the output. Everybody agreed on that; it was sort of a condition of the people joining us.

Lynch:

What about the group leaders? How did you pick up on Taylor, Richter, Mozley?

Panofsky:

Of course, again each one of them had a separate history. Taylor, of course was a holdover from the High Energy Physics Lab; Mozley had come with me to the High Energy Physics Lab from Berkeley mainly because I admired very much the way he was doing electronics. Actually he straightened out the control system problems at HEPL. There were actually very few people coming from the outside, most of the group leaders were simply parts of the original study group to build the laboratory. Even Richter joined us in the early days when there were openings at the High Energy Physics Laboratory and the Physics Department at the instructor level; I went East and did many interviews and recruited both Richter and Jerry Pine and we almost tossed a coin because these were openings of a different character: one on the instructor level with more teaching responsibility and the other one on a research associate level with more research responsibilities. So, it was just a matter of personal interviews and gaining confidence.

The thing which is quite remarkable in retrospect: there was really very little systematic procedure, and also in some respect very little democracy in the fact that there weren't any rules as to who had to approve what appointment. There wasn't any required consensus or voting by an organized body as to who would be appointed to be group leader and all that kind of thing. That was really a very personal matter but I did an awful lot of talking to a lot of people. It was basically a matter of interviews. During the early days of SLAC only very few people actually came from the outside: Perl came from the outside after a fairly formal search procedure. Before that people either were individuals whom I had worked with before or people who simply had been partners in the original deliberations which shaped the laboratory when we went from the High Energy Physics Laboratory to SLAC.

Lynch:

So Sid Drell fits in that category as well?

Panofsky:

Sid Drell very strongly fits in that category; he was an Instructor in the Physics Department in the early days. He went away for a year or two and came back as Professor. It certainly did not meet the modern model of a very systematic nationwide search. We didn't do that.

Lynch:

Whatever it was, it worked.

Panofsky:

It was basically a matter of personal chemistry; there was much of that.

Lynch:

You use the words personal chemistry; you gave me a logical lead into my next question. Apart from being well respected as a director of the laboratory by the physicists, I think I can safely say that the non-scientific people had a high regard for you. What sort of interactions did you have with the non-scientific people?

Panofsky:

Well, in a funny way I would get into their "hair" very extensively, again with the definite business that I would like to interact with them not as director but as sort of as technician. For instance, I would spend long periods of time with the engineers on the alignment system, and I would have long arguments on vacuum systems, and so forth. So they felt comfortable with the fact that I was a fellow "plumber" if you wish when it came to making decisions on technical systems and that they felt they weren't being judged by those informal interactions but that there was an honest, straightforward system of decisions. And that helps because that way they knew that I was interested in what they were doing but that I wasn't all the time conducting a performance review or something like that in modern terms.

Lynch:

Something which is a little unusual about SLAC, and that is that it has a tradition that the directors and the associate directions have a "life sentence" to the job. In retrospect, do you think that is a good idea?

Panofsky:

I know the present general custom is otherwise, is to appoint the director and deputy director, and maybe even group leaders, for fixed terms. I am not sure which is better; I personally participated in the University of California system where the director of the different laboratories under the University of California has five-year terms and then there is a very rigid review with an outside committee as to whether they should serve another five-year term or not; sometimes they don't want to anyway, but sometimes they want to. That has, of course, the great advantage that the director doesn't feel he's got tenure to the job. I emphasized frequently in faculty and other meetings, that faculty members have tenure but that directors do not have tenure. I was highly aware of the fact that I served at the pleasure of the President and that the President would get advice from the Scientific Policy Committee as to how the lab was going and the associate directors knew they served at the pleasure of the director.

I don't think anything much is gained by having a fixed term of review, but that's arguable. Anyway, we didn't do it, but at the same time we did, to compensate for that, frequently emphasized in meetings and other discussions that nobody in the administrative chain had a claim to the job for indefinite period of time. Since most of the technical associate directors were also members of the faculty, that was actually a good thing because it meant that they could be rotated without getting fired. In fact, we did rotate leadership in the research division with the individuals continuing in their technical and scientific role. So again, these things were not very systematic but we explicitly rejected the idea of term appointments for the directors. We discussed it and rejected it and replaced it by an explicit non-tenure arrangement.

Lynch:

When you say "we" what do you mean by "we?"

Panofsky:

I basically meant we had meetings called the "director's check-off," which were basically a very efficient way of running the laboratory. We would meet at a fixed schedule for several hours every week. By "we" meaning the Associate Directors and the Directors, and we would set agenda for these meetings beforehand, and we deliberately set agenda ranging all the way from very profound policy matters to other things which were complete trivia, but where it simply was more efficient to just decide the trivia rather than writing lots of memos and having it eventually come up the line. We would just talk it over and do it. In those meetings we discussed many things on a more profound level, as to fixed term service versus non-fixed term service and so forth, and we kept minutes and we set agenda, but we did not have any kind of process in which we would decide between the sublime and ridiculous as to what would come to the level of the directors. Basically any director or anybody from the outside could put anything onto the directors' check-off agenda and short of being forced to clip that off by saturation, that worked just fine.

Lynch:

Here's a somewhat provocative question: It is almost universally agreed, I am sure, that you did an outstanding job as director for a very long time. Is it possible that you did a dis-service to high energy physics by doing such a good job for so long that you made the job look easy?

Panofsky:

Oh, heck no. Apart from anything else, the job of being a lab director just has become very much harder during the last decade or so due to external and other factors, namely the erosion of trust between the governments sponsors and the universities and the general leveling off of budgets which has made things much more competitive rather than, if you wish, complacent during the growth of SLAC. To be frank, I think I was just plain lucky that over this long period of time the external conditions which are now causing so many crises in science management simply weren't there. That, together with the fact that we were very fortunate in having good people, and that there was a very fertile area of research really worked very well.

Lynch:

When did you retire and how did you decide when to retire?

Panofsky:

I decided very deliberately to retire as director at the chronological age of 65. We had a gentleman's agreement among the then directors to do that and I saw no reason to change that. In addition to that, I wanted to retire while I was ahead just because of the fact that things looked like they may becoming more difficult; this was the beginning of the era of storage rings. Burt had done a good job on SPEAR and SPEAR had been incredibly creative. So, I thought it would be much better to retire while the new director would not be burdened by initial crisis management but do whatever he felt should be done. I wrote a carefully worded retirement resignation letter pointing out that 65 is 65 and that the lab was in pretty good shape. I always thought it was very much better for a director to retire under that kind of circumstance.

Lynch:

That was just agreement between you and the associate directors at that time? That's not a SLAC tradition?

Panofsky:

No, it was an agreement among the Directors and the faculty and senior staff. It may have been discussed in the faculty, I don't remember, that bureaucrats should retire at age 65 and faculty members at 70. At that time, as you know, legislation was going through the Senate which made a requirement of retirement by chronological age illegal. So, universities or other organizations could no longer fix a mandatory retirement age. I don't remember now exactly when that came into force, but it was sort of in the wind for a long time, so we decided to make an internal, if you wish, not legally binding agreement of setting ages of retirement for bureaucrats and for faculty. Needless to say that agreement in the future did not totally hold.

Lynch:

What role did you play, or not wish to play, in the choice of your successor as director?

Panofsky:

I said explicitly in my letter of retirement to the President and Provost that I was retiring at age 65. I was announcing this a year-and-a-half, or whatever it was, ahead of the actual date to give them plenty of time to make a search for successor, and then I also said explicitly that in a list of candidates clearly Mr. Richter was a most logical choice and that if he were to be chosen this would leave a vacancy at the associate director level which again would require a search. So therefore, whichever way the President decided to go, there would be an extensive search procedure needed and therefore I wanted to signal that a year-and-a-half in advance. So, I did signal the fact that there was a clear candidate in-house in my letter of retirement, which is in the file somewhere.

Lynch:

Let's shift gears completely and go into the arms control business. In the previous interview in 1974 you talked about your arms control activities up through only 1965 and I know that arms control activities have been a large part of your life, both during your tenure at SLAC and after you retired. Let me ask first, I know that you have been on several different National Academy panels, what were some of those.

Panofsky:

Well I was initially on lots of committees that were not National Academy and then some of them later became National Academy. Probably my first formal committee assignment was on the Scientific Advisory Board to the Air Force. This had to do with the establishment of early warning systems just at the beginning of the cold war, whether some early warning system should be set up north and what use it would be, and if there was warning what would you do with the warning. Then I independently joined what was then called the Northern California Association of Scientists, or something like that. It was basically a predecessor to the Federation of Atomic Scientists, making yourself available to make speeches about nuclear energy. The Scientific Advisory Board of the Air Force was my first formal assignment.

Lynch:

When was that roughly?

Panofsky:

That was when I was still at Berkeley. It must have been '48 or '49. That I think was my first formal assignment, in addition to the various lectures, I don't really remember the history very well. There followed a formal review, when the initial nuclear test ban discussions started on the question whether a test ban could be violated by nuclear testing in space. I was asked by the White House to chair a panel to look at the technical potential of evading a test ban by testing in space. I thought that activity was very important because it tried to grapple with the fact that verification of an arms control agreement should not be an absolute requirement. We established a standard that if the effort to evade an injunction of an arms control agreement was so large that our national security might be better served if an opponent would put his resources into evasion of an arms control agreement rather than doing something else which was hostile to our interests, that maybe that was just as good.

I think we did a good job to counteract the rhetoric that if anything is non-verifiable that's bad without putting some kind of quantitative measure on that. After a report was issued, which was generally accepted over a very broad spectrum of opinion between hawks and doves, I was asked on very short notice to chair a delegation to talk about detection in space with the Russians in 1959. The reason for that — that's all very well known history — is that in 1958 there was a thing called the Conference of Experts where the American chief delegates were Jim Fisk and Hans Bethe. Then there followed two technical working groups to settle unresolved questions along those lines. One was on detection in outer space of possible nuclear tests, which I chaired, and then another one on new developments on detection of explosions underground, which Jim Fisk chaired and I served as his deputy. That was a large organized activity.

I then became a member of the President's Science Advisory Committee and its agenda were very heavily loaded with arms control issues. At that time there wasn't any mechanism in the United States government to deal with technical issues related to arms control, and in fact one of the origins of the Arms Control and Disarmament Agency was that the President's Science Advisory Committee said that our agenda are so overloaded with arms control issues, you have to institutionalize it and get it out of the committee. Then I continued after my membership terminated, there was a fixed term — I forgot what it was, 4 or 5 years — I continued as a member of the Strategic Military Panel of the President's Science Advisory Committee for some time. Then, I was involved in several specialized alternate basing studies for the Office of the Science Advisor and I was asked by the Navy to look at alternate choices in air-to-air missiles between the Sidewinder and more complicated commercial systems which were then produced. Then I was involved in another panel on infrared detection of missile launches. I was also consultant for the Arms Control and Disarmament Agency and participated in some organized studies, mainly on peaceful use of nuclear explosions.

There are a whole bunch of other assignments, most of which were not connected with the Academy. On the Academy side the following thing happened. Before the National Academy of Sciences got into arms control very heavily, the American Academy of Arts and Sciences under the leadership of Paul Doty from Harvard University started quite early to engage in bilateral discussions on arms control issues with the Russians. Doty, whom I knew from PSAC days roped me into that. Then the Russian counterpart — Millionchikov in those discussions died and Doty and I decided it would be good to transfer those activities to the National Academy of Sciences from the American Academy of Arts and Sciences because they gained higher visibility that way. Then we persuaded the National Academy of Sciences to institutionalize an organization which is called the Committee on International Security and Arms Control, which should have a continuing role in arms control. Now that was an innovation for the National Academy. The National Academy usually appoints panels ad hoc to address a given science and policy issue rather than having a standing group which continues to cover a given area. Paul Doty and I persuaded the Academy that this issue was sufficiently important to be institutionalized.

Lynch:

When did this happen?

Panofsky:

I would have to look that up. I would say in the late '60s but I would have to check the record. Doty became the first chairman of this group and I was a member throughout this period, then Goldberger became the chairman, and then I became the chairman and now John Holdren is the chairman. It has been a very vigorous activity of the National Academy. At various times I got involved in other organized committee study activities. I recently chaired a committee to review the research program of the Department of Energy for remote sensing and other technologies for early warning on proliferation activities. Also, I was requested to testify at various times in connection with these things. I played a relatively major role during the 1959 debates on whether the so-called Safeguard ABM system should be deployed; there are lots of stories about that.

Lynch:

That Frankenstein monster kept coming back.

Panofsky:

That's right. I don't know if you want to talk about the famous story of my encounter with then Deputy of Defense Dave Packard.

Lynch:

Sure.

Panofsky:

What happened was that at the end of the Kennedy Administration Kennedy decided to deploy a limited ABM system, called the Sentinel System, which was supposed to protect against accidental ballistic missile attack, and McNamara made a famous speech about that. Then when the next Administration, now wait a minute - that was under Johnson I think, not Kennedy. It was McNamara's speech in Ann Arbor which announced the so-called Sentinel System. Then the next Administration, with Dave Packard being Deputy Secretary of Defense, decided to convert that system from a thin-defense over the country to a point-defense for strategic missile sites in the United States but using the same hardware. The amusing thing was that I ran into Dave Packard at the airport and I went with him to one of the airline lounges and he asked my opinion about that. I told him that it was my opinion that the mission of replacing area defense by point-defense was actually a good thing and stabilizing but that he had the wrong hardware. So we talked about it; nothing happened. Then I was several weeks later in Washington to testify before the Joint Committee on Atomic Energy about SLAC.

I noticed that there was a hearing about this new ABM system in the next hearing room and I went over there to listen after my testimony was completed. Packard was testifying and he was asked, did you have scientific advice in your decision and after a while he said, yes, I talked to Dr. Panofsky. The Senator asked him, where did you talk to him? and he said in an airline lounge at the airport. That caused a major uproar and caused a column by Arthur Hoppe, the syndicated columnist, as to how the Pentagon gets scientific advice. Then they asked me to testify on the subject independently and I told Congress the same thing which I told David Packard, namely that I approved of the mission but not of the hardware. Later, the whole ABM business became a full-blown debate and I was asked to testify quite a few times and solidified my own views on it. I spent a fair amount of time both writing and giving testimony, writing some articles and so forth.

Lynch:

That was a long-standing thing; originally it was Safeguard and then the SDI.

Panofsky:

Well, it's a very complicated issue and this is probably not the right time to air it. There is confusion, this being a complicated issue. If you defend the land-based arm of the deterrent forces that presumably improves the survivability of the land-based deterrent forces and makes better deterrent forces. But if you defend the whole country, then you make it more difficult for the opponent to maintain his deterrent, so the opponent would tend to escalate. And if it is cheaper for technical reasons for the opponent to defeat the defense, then you cause an arms race which simply makes you no more safe but simply escalates potential lethality and escalates cost. So in a paradoxical way area defense for technical, not political reasons, is destabilizing, while point defense of the strategic deterrents is not. So that issue was being debated all over, and of course it tends to be distorted into "how can you be against defending the country" kind of simplistic argument. I tried my best to counteract that. As you know the Safeguard system was passed by Congress by one vote and then was canceled very shortly thereafter, simply because of the high cost relative to what it could do. I stayed heavily involved in that issue ever since.

Lynch:

Some of the international discussions on international control has been with the former Soviet Union. Such discussions are often face-to-face discussions. Who were your counterparts there? With whom did you interact?

Panofsky:

That actually was, I thought, one of the most constructive things. We first started on bilateral discussions with the Russians and we agreed on the dynamics that they were off the record, we would not agree on anything, there would be no agreed positions or agreed proclamations but that each of us would discuss things and each of us could and would freely report to our governments but yet we were not negotiating. People on the Russian side were such prominent people as Academician Velikhov, and Basov — the co-inventor of the laser, Prokerov, Sagdeev — who is now in this country, and Kapitza; senior academicians.

Lynch:

Is it Kapitza, the younger?

Panofsky:

The older and the younger. The older in the old days and then later the younger, several others — that's all in the record and we would carefully document unilaterally what was discussed but we would not in any way have any jointly agreed minutes let alone declarations, because that would clearly lend itself to abusive propaganda. The idea was only to have a channel of communication. Many matters were discussed in those bilateral discussions which much later came on the official agenda. For instance cut-off of fissionable materials was discussed in the early days, and I wrote some papers on that, and of course ABM issues and the whole question about the causes of the escalatory arms race, and things of that kind. Those things were very valuable to the Russians, but the Russians of course had to clear their prepared talks. These meetings were valuable because when the Russians said something new it meant that the government might be inclined to be flexible in the same direction. So it had some real value of fathoming Russian official intentions.

Today when you meet with Russians, the opposite is the case, they are totally unimpeded and talk nonsense, just like the Americans, and therefore from the human point of view these discussions are much more satisfactory because they are not constrained. But from the value of sounding out official positions they become less significant. Then we decided in the arms control discussions at the incentive of David Hamburg to explore whether the Europeans would be interested to create arms control groups in their academies; we felt these bilateral things with the Russians were so important that it was sort of paradoxical that we would talk to our adversaries and not to our friends. So, I took an initiative and I went around and gave talks to the French Academy, to the Royal Society, the Italian Accademia de Lincei, and persuaded them to establish similar standing committees as our Academy did. That took root and is now quite active and successful in the UK and in Italy, and to a lesser extent in France. It has given birth to the so-called Amaldi Conferences which are a round-robin of all European Academies talking arms control. That goes on every year and we took the initiative for that.

Lynch:

How important do you think it was to have respected physicists on both sides involved in discussions as opposed to weapons specialists or State Department officials?

Panofsky:

It's not a question "as opposed" but it's a matter of "and." I think they are very complementary. Clearly it was very important to have independent academic people doing that for a simple reason. Academic people tend to be more respected by governments as being independent and not being motivated by personal gain or institutional interests in talking about military matters. At the same time it is also clear that academic people talking to one another form a certain community of interest and they can approach problems in a problem-solving spirit but they cannot commit their governments. So whatever they say can be, and often is, totally discounted by governments and doesn't get anywhere. So, the value is that this is an informal channel. Since scientists are usually more listened to on technical matters by their governments than laymen, there's certain credibility to that but it's in no way a committing channel. So, it has value but it doesn't replace discussions on an official level.

The problem is that people are always looking for the "independent expert." If you talk about such matters as weapons, clearly the academicians are not as expert as people from the professional weapons laboratories and so forth. On the other hand the people in the academic community are more independent. So the main thing which I felt very strongly about is that we need the independent discussions, discussions by independent scientists, but that we should make a positive effort to have the independent scientists become more expert. Therefore in the standing committee of the Academy dealing with arms control, we have the opportunity not only to talk to one another but also to do the homework by doing independent studies advising the government so that to the maximum extent possible they could combine the independent expert ideal rather than being one or the other.

Lynch:

Was that difficult to achieve?

Panofsky:

No, but it took time to achieve; it wasn't difficult but it couldn't happen instantaneously. On the other hand it is much easier to achieve this in the United States because there are so many people in the academic community who had been involved in military affairs during World War II who then went back to the academic life but still maintained some competence. But of course that generation is now disappearing. One of the purposes of the Academy activities since happily there has not been a war involving the United States on a large scale for over 50 years is to make a positive effort to produce a new generation of independent scientists who are reasonably expert on military matters but who are at the same time not institutionally or professionally making a living of military matters. For instance, the Academy group and the JASON group and several such other such groups have that function to introduce younger people into the, if you wish, independent expert category on these matters.

Lynch:

A Comprehensive Test Ban Treaty has been discussed for many years, but some major changes have occurred recently in this area. Did you play a role in these changes?

Panofsky:

No. I did not play a role in any direct way in the recent achievement of the nuclear test ban treaty. I did play a role — it depends on how far back you go — I discussed both in Congressional testimony and other studies the threshold test ban treaty which was signed quite early but not ratified for over a decade. There the issue was with what precision can you measure the yield of nuclear explosions. The threshold treaty bans nuclear explosions over 150 kiloton equivalent in power. I always was very worried about it because if you have a quantitative cut-off like this, then allegations of cheating would always be made simply based on bad measurement. Therefore there was a big debate, the Reagan Administration taking the position that you needed very high precision to measure the threshold.

Then there was a technical debate as to whether the threshold should be measured by tele-seismic means, that means measuring seismic waves at long distance versus so-called hydrodynamic means where you would have to go close in with measuring instruments to determine how the velocity of sound is affected by amplitude when you come in very close to an explosion; here the sound propagation becomes non-linear and the velocity depends on the amplitude. But to do that you have to adopt much more intrusive methods. There was a big debate on the relative merit of hydrodynamic versus tele-seismic measurements and I played some role in that both in testimony and in analysis. But as far as the Comprehensive Test Ban is concerned, I participated in some public statements certifying to the fact that I believed that safety and security of nuclear weapons would not be dangerously impaired under a comprehensive test ban.

I participated in a JASON study which Sid Drell chaired on whether a Comprehensive Test Ban Treaty could be truly comprehensive or whether some relatively small so-called hydro-nuclear explosions should be permitted under a test ban. In more recent days I have always been a supporter of the test ban in various public and other appearances. However I also felt that opponents and proponents, both over-stated the importance of a test ban treaty from a technical point-of-view. A total test ban treaty has an enormous political significance in support of non-proliferation, but technically you can develop nuclear weapons without testing as South Africa has done, and the first Hiroshima bomb was never tested. So, it doesn't stop proliferation. Tested weapons can be manufactured in any arbitrary quantity so it also doesn't stop accumulation of nuclear weapons. So technically speaking a Comprehensive Test Ban doesn't control either horizontal or vertical proliferation. But the inverse is also true, namely whether verification is very good or not it is also not a major factor for national security. I always felt in recent times, in the test ban debates, that both proponents and opponents were overstating their respective cases, although I am a proponent within that limitation. It is extremely important politically but less important technically.

Lynch:

What role do you think technical issues versus political issues play in moving towards the test ban treaty.

Panofsky:

Oh, I think in recent times they play a major political role because a nuclear test is after all the largest man-made event in terms of energy release which people have done, so therefore whether you can get such tests under control symbolizes politically whether technology controls man or man controls technology. It always has had very big political significance, and of course it has an explicit political significance because in the Non Proliferation Treaty, in particular in the recent renewal of the Non Proliferation Treaty for an indefinite duration, achieving a comprehensive test ban treaty is explicitly referenced as being an achievement which must be made to signal that the participants are adhering to the treaty in good faith. So it has become politically extremely important, both in terms of its broad symbolism and in terms of its specific signaling adherence to the Non Proliferation Treaty. My earlier remarks, that from a purely technical point-of-view the leverage the Comprehensive Test Ban Treaty on arms control is not very large, simply means that the political considerations tend to be dominant and maybe should be dominant in discussing the issue.

Lynch:

What role do you think the collapse of the Soviet Union played in the change of attitudes towards arms control agreements.

Panofsky:

I don't think really very much. I don't think, with the benefit of hindsight, the rate in arms control progress and the attack against arms control progress has changed very much after the collapse of the Soviet Union. The collapse of the Soviet Union induced arms control issues of their own, namely who is the successor state of the Soviet Union and the problem that once the Soviet Union collapsed you suddenly had four nuclear weapons states rather than one. So, the arms control issue which ended in the Lisbon Agreement of moving all nuclear weapons in the states of the former Soviet Union to Russia, that was an example where the collapse generated arms control issues on its own. On the other hand, the arms race has been reversed and shifted from concern with all out conflict among the nuclear weapons states, to proliferation, to the problem of proper maintenance and guarding and safeguarding of nuclear materials and so forth. So, the focus has shifted and unfortunately the collapse of the Soviet Union and of the cold war has given a sense of false security in my view. Even though I think the risk of a nuclear holocaust has decreased, the risk of something bad happening in the nuclear area, if anything, has increased. Risk is the product of probability and consequence. So I think the maximum consequence has decreased but the probability of something bad happening has increased.

Lynch:

Yes, could you tell us the babushka story?

Panofsky:

Well, one of the main issues which has come to the forefront is that in the old Soviet Union the control and management of nuclear weapons was in the hands of highly trained and highly disciplined Soviet troops. There was relatively little back-up behind that in terms of accounting systems for nuclear materials and what is known in the trade as management, protection, control and accounting (MPC&A). Now with the discipline essentially having disappeared and troops being underpaid and partially demoralized there is of course increased concern along those lines. The Committee on International Security and Arms Control organized a study on the management of plutonium which is withdrawn from nuclear weapons as a result of arms control, and in that connection we of course became impressed by the relatively deficient management, protection, control and accounting on the side of Russia. You ask me about the babushka story in connection with that work; I had a discussion with the head of the Nuclear Regulatory Commission of Russia. The gentlemen said that the protection of our nuclear material has shifted from highly trained troops to babushkas (grandmothers) wielding a cucumber. This did not inspire much confidence.

Lynch:

And some of the grandmothers did not know how to handle the cucumber.

Panofsky:

And the grandmothers did not know how to handle the cucumber very well. That of course is a widely recognized fact now; there has been much progress in strengthening that control, mainly as a result of laboratory-to-laboratory activities directly between American weapons labs and Russian weapons labs, but there is a long ways to go. So I certainly can't answer your question whether the collapse of the Soviet Union should make you feel any safer or not at this particular point. It has affected arms control in some positive ways and in some negative ways. It created more issues in arms control but it has diminished the interest of the average citizen in arms control, simply because they think there is no problem and that is not true. So making progress in arms control is very difficult and when there are debates on arms control, as we see now in the Congress, it is very difficult to get the public to pay any attention to them.

Lynch:

You mentioned the plutonium problem. I know that you were part of a committee that was looking at plutonium disposal. Can you tell us a little about that.

Panofsky:

Yes, that was an interesting experience. Actually it was not a new committee but the standing committee of the National Academy on International Security and Arms Control taking on this task. The history of that was very interesting. The history was that in our bilateral discussions with the Russians, we discussed the plutonium question and we came back and I briefed Brent Scowcroft, at that time the National Security Advisor to President Bush, about it. Scowcroft said; "Pief, the United States has no policy on this issue. Could you guys do a study on it and I said yes. That was the beginning of the study; Scowcroft then instructed the Department of Energy to support the travel costs and what have you to do this study.

So we did this study and we found that the Department of Energy had a policy which was actually adopted under very much pressure from Congress, which Congress had adopted under pressure from reactor vendors, that the way to deal with the plutonium issue is to build a whole new generation of special reactors, specially dedicated to use the weapons excess plutonium and burn it up in these new reactors. We came to the conclusion after very extensive deliberation, a total of about a year-and-a-half, that there was no need to build new reactors for this purpose. There were two classes of plutonium, that was well known; namely excess plutonium withdrawn from nuclear weapons and the plutonium which was now in spent fuel from commercial nuclear reactors.

The spent fuel plutonium was not a particular risk from the proliferation point-of-view because the spent fuel is so radioactive that it is extremely difficult to clandestinely steal it or divert it. So we felt that the nuclear weapons withdrawn plutonium need not be sequestered any better than the commercial plutonium contained in spent nuclear fuel. So our recommendation was to use existing reactors, make special fuel which would be burned in commercial nuclear reactors. If that was not practical, to mix the plutonium with highly radioactive waste and put the whole thing together in either ceramics or glass and dispose of it that way so that its radioactivity would protect it against diversion and theft. After this recommendation was made the United States government adopted it essentially exactly, but it was opposed, interestingly enough, from three different directions: firstly from the reactor industry who didn't like it because the reactor industry, which currently is very hungry because there has not been an order for a new reactor for a long time, hoped to use our concern with plutonium to leverage acquisitions of new reactors and we basically felt that this was not necessary.

Then it was highly opposed by peace and environmental groups who were opposed to use plutonium in any constructive way because that would signal that plutonium was a useful fuel in the future even though right now the purpose was to get rid of it. Finally, we were opposed by groups the other way around from the reactor industry and also by the Russians who didn't want to throw away the energy content of the plutonium. So here was an interesting case where we came up with a very specific recommendation which we thought was very constructive and the United States government was largely persuaded to follow that line, and on the technical level so were the Russians. But the strong opposition which still remains unresolved came from the people who had a different agenda, namely the future of nuclear power, the future of the reactor industry and so forth which they felt was more important right now than arms control. So this is part of the problem, that with the dissolution of the Soviet Union, arms control is considered by some to be considerably less important than some other values or agenda which they hold dear.

So if you want to make any progress on arms control you tend to run into opposition from people who consider basic arms control objectives to be less important than some other values which they are trying to promote. So I found it a thoroughly interesting experience. On the one hand I have never been involved in a committee which was as effective in having its recommendations accepted by the government, but at the same time it was somewhat disheartening to have so much opposition from people who had other, if you wish "fish to fry," which may be important but which in my view are not as important as carrying on the original arms control objective.

Lynch:

Let's shift gears and do something different. We talked about arms control; another thing which you were involved with outside of SLAC and now after your retirement is the SSC - the Superconducting Super Collider. Just for the record can you say briefly what the SSC was and what role it was supposed to play?

Panofsky:

Yes. Well, as you know the energy and general capabilities of accelerators has been steadily increasing; roughly speaking the energy of accelerators in the laboratory has been increasing by a factor of 10 every 7 or 8 years. That progress had been made possible by the fact that the unit costs, the costs to achieve that energy has been decreasing so that the cost of any given accelerator installation has not been going up anywhere near as fast as the energy. Periodically the community of high energy physicists, either through internal committees or by committees appointed by the government, has been reviewing what the next logical step should be in accelerators and colliders.

The general agreement is that electrons and protons were complementary in solving problems in elementary particle physics, and the last decades or two have been extremely productive in giving a coherent picture of how elementary particle physics works in the so-called Standard Model; this is one of the great intellectual achievements. But the standard model has simply too many parameters which one has to feed into it externally. There clearly have to be physical reasons for those parameters and they can only be accessible at higher energies. So the various deliberations on this subject resulted in the conclusion that a machine which could have the general order of 1 TeV energy in the frame of the colliding fundamental constituents would be the next logical step. In terms of proton energy, having 1 TeV order of magnitude collision energy among the constituent quarks within the colliding nucleons would mean a proton machine of around 20 TeV per beam collision energy.

So various successive review committees therefore made that type of machine the next logical step. A design study was done at Cornell University, which made a very rough cost estimate and a rough estimate of engineering feasibility of the machine. Following that very preliminary analysis, a formal action was taken in which the University Research Association (URA), which is a consortium of universities which has management responsibility for Fermilab and supported by the Department of Energy, made a proposal for the construction of a machine for proton-proton collisions of 20 TeV. That was called the SSC. That machine had certain other requirements as to data rates which had to be produced because at these high energies the basic cross-sections go down. So all this culminated in a proposal which was 20 TeV against 20 TeV protons and a projected luminosity of around 10³³cm^-2sec^-1. In the first proposal it was decided for political reasons to submit the basic proposal for such a machine based on a site independent feasibility analysis, in parallel that with a site selection which was to be done by a preliminary screening by an Academy committee and then final decision by the Department of Energy. So all that is sort of a thumbnail sketch of what the SSC is.

Lynch:

What role did you play in the SSC?

Panofsky:

The arrangement was that the proposing and then the contracting entity, University Research Association, has a Board of Trustees. Then it was decided that since the responsibility of URA would be split between Fermi Lab and the to be SSC, that there should be two so-called Boards of Overseers which would report to the Board of Trustees of URA; one Board of Overseers for Fermilab and one for the SSC. I was appointed to be a member of the Board of Overseers for the SSC and then, following Boyce McDaniel, became its chairman. So I had the responsibility for reviewing the proposal preparation. Once the project in its initial phase came into being this responsibility extended to the actual research and development leading to the machine. That was an extremely difficult process. The design was initially pursued by a very able group headed by Maury Tigner, called the Central Design Group, which was supposed to design the machine in a site-independent way and refine the cost estimates. Then the wheels ground and the site at Waxahachi, Texas was selected and actual detailed site-dependent design was done and I was involved in my role as chairman of the Board of Overseers to oversee that. We had periodic meetings and we were in the midst of very stormy problems which beset the SSC.

Lynch:

One of the characteristics of the SSC was that it was so large that it needed foreign participation. How effective was the U.S. effort to get foreign participation, and why?

Panofsky:

The U.S. effort in getting foreign participation was relatively poor for a number of reasons: First, the United States government was internally quite un-united whether to even seek foreign participation. One of the reasons was that, when the Executive branch of the government was persuaded to support the SSC, it was initially promoted by President Reagan in announcing SSC to be an example of American competitiveness. There is of course a direct clash between competitiveness on the one hand, and international collaboration to share costs with other countries on the other. You can't have it both ways. Another thing which was sort of interesting was that the Executive branch commissioned a study to investigate whether international collaboration of the SSC would engender too much technology transfer to foreign countries which would then endanger either industrial competitiveness or even possibly military security. One amusing thing was that that study was classified.

This study warned about all sorts of things which would happen if there was international collaboration or too much transfer. At that time the then director of SSC was not allowed to read it. I had the privilege to read the report and found that it was technically extremely ill-informed. For this and many other reasons it was not at all clear that international collaboration was even desired, let alone promoted, by the U.S. government. As time went on, the U.S. government in the Executive branch did agree that an international collaboration was very desirable, but they still didn't support it with very high priority. For instance, we through the Board of Overseers and also scientific spokesmen urged the Executive branch to give high priority to the SSC in discussions at summit meetings between the Japanese Prime Minister and the American President, but that never happened because other things had higher priority. So on the Executive branch side there was a shift from competitiveness to collaboration and I was somewhat instrumental in a session with the Secretary of Energy to push that. But, on the Congressional level, in debating the wisdom of having the taxpayer pay for the SSC, international collaboration was very heavily pushed because it presumably would decrease the bill to the American taxpayer. So even from the very beginning there was always a bit of dissonance between the two branches of government as to how important the international collaboration really was.

My personal view is that an international collaboration is extremely valuable for two reasons: intellectually, because it takes an international effort in the community of scientists to really use a machine of this kind, and also to signal the international character of science. But then it would also save money to the American taxpayer, but not very much because when you run things on an international basis there is some unavoidable decrease in efficiency because you make the decision-making somewhat more difficult. Also the United States, after all, has about one-third of the GNP of the entire world, so the factor which is involved in terms of potential cost-sharing can never be very, very large. So international collaboration is an extremely important thing to achieve but one shouldn't have any illusions that when a machine gets built in the United States that it would suddenly become very inexpensive because it is international.

Lynch:

Yes, and unfortunately the SSC was not successful and it was canceled by Congress in 1993. What problems led to the termination of the project.

Panofsky:

Well, it's a very complicated issue. I have written on the subject giving my personal views. A formal study on that is in progress sponsored in part by the National Science Foundation. I found there is a fair amount of responsibility, or guilt if you wish, to go around. Firstly, a new Congress was elected and most of the negative votes against the SSC came from the freshmen members of the Congress who simply felt instructed not to vote money for something they didn't understand and which did not demonstrably benefit their constituencies. Neither the Executive branch of the government, nor the scientific community did a very good job in really explaining to the body politic and the public what the benefits would be. So that was one reason. The other reason was the fact that the support by the scientific community for the project was good whenever there were committees convened to look at the value of the SSC.

However in terms of voting with their feet, namely actually joining the staff of the SSC and working hard and constructively on the machine, the cooperation of the scientific community was not that great and recruiting was very difficult. One of the reasons why recruiting was difficult had to do with the fact that the Department of Energy would over-administrate and over-micro manage the laboratory. A problem was that the leadership of the Department of Energy, starting with Adm. Watkins and his successors didn't have confidence that, notwithstanding the excellent record of members of the scientific community managing these large laboratories in the past, in the case of SSC that they could do it in the future. Therefore the DOE created, in my view, a monster organization of over a hundred people in the site office of the DOE which was to oversee in a direct way the building of the accelerator, so that not only the management of the project, but also technical group leaders, had to spend an inordinate amount of time in continuously reporting their work to various levels of DOE at the local site office.

That in turn contributed to spoiling the atmosphere at the laboratory in the sense that procedures were highly bureaucratized and that line responsibilities were diluted by having both to report to the responsible people at the laboratory but also all the time, even in relatively minor matters, to DOE. There was budgetary control on an extremely large number of line items; any changes in cost of these items had to be approved by DOE. It is impossible to gather the needed technical competence in this type of an external DOE oversight organization. That in turn, call it bureaucratization of the lab, made it harder to recruit people because it simply was not as much fun to work under those particular conditions. And one, in retrospect, can criticize both laboratory management and the Board of Overseers of not resisting strong enough this kind of micro-management encroachment. If you oppose governmental micro-management too much, you run the risk of the project being canceled. If you don't oppose it, then you impair the productivity of the project. But the way it ended up it got the worst of both worlds, namely the productivity was interfered with, recruiting was more difficult, but it got canceled anyway. However the basic reason for the cancellation, in my view, was just plain budgetary. The Congress simply was unpersuaded that a basic science project of this magnitude should be funded.

Lynch:

What do you think the long fall-out of the cancellation of the SSC is on the high energy physics program as a whole?

Panofsky:

It is hard to predict the fall-out from the cancellation of the SSC. It's very hard to separate cause and effect: there is general pressure on basic science today and whether the cancellation of the SSC was part of that or whether the cancellation of the SSC has emboldened people who are opposing support of basic science is hard to disentangle. It's certainly not good. What happened was that a machine one-third of the energy of the SSC was at that time under deliberation and planning in Europe, the so-called LHC. It has now been officially approved in Europe. The government's High Energy Physics Advisory Panel recommended strongly that the United States should become a participant in and contributor to that European machine, but right now those U.S. contributions and participation are very much in limbo, again as a result of the Congressional debate which is now ongoing.

In addition substitution of the LHC is bad for science because the LHC energy is a factor of 3 lower and the certainty that the LHC will uncover evidence for the parametrization of the Standard Model is much less than the certainty that the SSC would have done that, simply because of the various physical limits which pertain to that kind of thing. At the SSC energy it was a virtual certainty that one would have discovered the reasons why different particles have different masses within the Standard Model. Under so-called unitarity limits such reasons would have to appear. At the energy of the LHC there's a very good chance that that can happen but there is no physical certainty that the evidence will appear. So it's bad for science, it's a bad precedent that the scientists were not able to convince the Government (a) that the physical necessity was there and (b) that the past management scheme which had been successful at SLAC and at Fermilab could not be continued into the higher energy regime of the SSC and that more oversight by external lay-bodies who really weren't very well qualified was not needed. None of these things are good. How destructive they are is hard to tell. And of course one may argue that one simply expected too much in spending that much money at that scale without that kind of oversight, even if you feel that such oversight is counterproductive to efficiency of the doing the project. So, it may be that way in the future.

Lynch:

Do you think the cancellation of the SSC will make future collaboration with foreign countries on other large projects more difficult because we are seen as an unreliable partner?

Panofsky:

I don't think so. I think the problem about the United States being, to use your word an unreliable partner in foreign scientific collaboration, that problem has been around before the SSC. It didn't enter into the general consideration with the SSC. Several other international collaborations in science have failed because of the fact that in the United States appropriations are being made annually. It is extremely difficult for the United States government to make a long-range commitment to an international scientific project. This is an indigenous problem; it can be solved either by extreme good faith or by having some very long-range contractual commitments, but the SSC is not the first and probably not the last where this whole question of how the United States can be a reliable partner in an international venture can be put to the test. Currently, as you know, there are major problems, whether the international space station, which scientifically has much less merit than the SSC, shall or shall not continue as an international venture. The kind of Convention that established CERN where governments really delegate responsibility to members of a Council is very hard to do here. In the United States it is very hard for the U.S. to delegate as much long-range fiscal authority to a representative as is needed for an international scientific venture. So it is a fundamental structural problem. The SSC surely didn't help but it was not a new item in respect to raising that particular problem.

Lynch:

You've raised several of the things I wanted to ask next, but let me ask the question, what other lessons do you think we can learn from the SSC?

Panofsky:

I think the lesson which I feel very strongly about is the following: If management, that is basically scientific management of a laboratory, feels that interference by the sponsor in the management of the laboratory is excessive then the management should be aggressive in opposing that and take more risk in getting the support canceled. In retrospect with benefit of hindsight, I believe that SSC management, including the Board of Trustees and Board of Overseers, was too timid in opposing some of the clearly bad practices which were being imposed by the government. More brinkmanship, if you wish, should be exercised on that. At the same, I think the scientific community did not do a good job in dedicating manpower to educating the public as to the merit of the work they were trying to do. You simply cannot expect public support of something of that magnitude without spending a greater effort in explaining it to the public. They certainly did not do a good job. Then, there was also the fact, although there are explanations for it, that members of the scientific community were quite willing to advocate that somebody should build a machine without putting their personal effort to it on the scale that was indicated. So the fact of the dichotomy between what you advocate and when you vote with your feet is also a lesson which one might possibly draw for the future. So, as I said before, the principal reason was that we simply did not succeed to persuade the public and the Congress that a single basic science project of this magnitude should be financially supported. On the next level there's lots of blame to go around and certainly that includes the Board of Overseers, Board of Trustees, scientific community, the DOE managers, and the DOE secretary in terms of mistrust. These things are all factors and I would not be willing to say that any one of these was really predominant in causing this tragedy.

Lynch:

One of the things you mention was that the physics community did not succeed in convincing the public at large, and the Congress in particular, of the importance. What do you think they could have done?

Panofsky:

Well, they certainly could have organized more personal discussion groups with members of Congress, and in particular Congressional staff, and personal visits with Congressional staff. They could have had more public foray in various communities. There is no one bold master stroke one can do, it is just a matter of doing an awful lot of work and mainly dedicating time. And of course some people of the community are better at this than others; some people are very good at explaining things publicly; some people tend to exude some arrogance and have the "father knows best" attitude and that is counterproductive. By and large I found that members of Congress are extremely accessible to scientists, much more than they are to commercial pressures. Most members of Congress feel in some respect honored when a scientist with a high reputation makes an appointment and just talks shop. That kind of thing was not done frequently enough. There were various specific convocations which were organized. Another problem which I failed to mention was that of course that somehow or other the project became regional. It was really very depressing that before the site was selected and there was competition for the site, a substantial fraction of the different states made proposals stating why (a) this was the greatest piece of science which ever existed, and (b) it should be in their state. But then once it was finally decided that it should be in Texas, then it became the Texas project and it got involved in the various competitive resource requests by different states. The scientists in the other states didn't do a very good job in persuading their representatives to say, "Look, you thought this was great science as long as it was in your state, how come you are now opposing it?" The thrust of that kind of argument was lost.

Lynch:

Let me wrap things up with your retirement now. You have been formally retired for some time but you are clearly very active; you even have an office at SLAC. Besides arms control activities we've discussed, what do you do?

Panofsky:

Basically, when I retired at age 70 I agreed with Burt Richter, my successor, to be a zero-cost consultant, which means that I continue to have an office and run up a telephone bill and so forth. I have been mainly engaged in external arms control activities; I also talk shop simply with members of the staff in case there are some problems. I have been doing some calculations on some orbit dynamics and beam detection problems, but by and large my activities have been outside the laboratory either in arms control or program reviews in high energy physics in other laboratories, or program reviews of technical but not exactly arms control activities. The arms control activities have been intensifying partially because of success of internationalizing the role of academics in arms control; that simply means more communication and more bilateral discussions, and writings with Chinese and Ukrainians and Italians, and so forth. So by and large I am still, by the rules of Stanford University, an Emeritus Professor who can be heard but not vote in faculty meetings and that I've been doing. In faculty meetings of SLAC we discuss both new appointments and alternate futures for the laboratory and so forth, so I still have a voice in those discussions. I have not been doing anything which you could remotely call research in particle physics. I have been writing some review articles on accelerators.

Lynch:

Let me ask the question in a different way just to get a different perspective. What are your pleasures in retirement?

Panofsky:

I'm actually a very poor retiree, in the following sense that I don't have very intense hobbies; it makes absolutely no difference to my economic well-being whether I show up at work or not. The pleasures are basically that I can do what I want to do. A pleasure is to be still alive! I've never had, even though I enjoy mountain climbing and traveling around, any intensive hobbies, so my hobby is just plain to continue. So it really is not a very large discontinuity, because even before retirement it is clear that in respect to active research, one loses competitiveness with the younger generation anyway. There really hasn't been that much discontinuity. I became more progressively involved in public affairs and administration anyway, even before retirement.

Lynch:

Now the ultimate good news-bad news question. What do you consider to be your greatest achievement and your greatest disappointment?

Panofsky:

I don't know. Probably the greatest disappointment has been the failure of the SSC because I've had the good fortune that nothing else really went totally sour which I have been involved in. Various things I have been involved in have been very productive and of great consequence, and some things have been more or less routine, so as far as disappointment is concerned, I would say without question, without ambiguity, the SSC. Achievements, probably I would say the early experiments on pion absorption in hydrogen and deuterium because that single series of experiments really had large consequences in terms of understanding the intrinsic parity of the pi, pion masses, and many other things. That was really a very major thing. Then of course comes SLAC, which is clearly a good thing.

Lynch:

Thanks Pief.