Oral History Transcript — Dr. Robert Frosch
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
See the catalog record for this interview and search for other interviews in our collection
Interview with Dr. Robert Frosch
Robert Frosch; October 6, 1981
ABSTRACT: Reviews Frosch's (b. May 22, 1928) education at Columbia University (PhD, 1952, theoretical physics) and, in detail, his varied career as a physicist and a science manager, beginning with his work as a research scientist at Hudson Laboratory (1951–3) and then as Asst. Director and Director of the Theoretical Division (1953–63). In 1963 he became Director of Nuclear Test Detection, Advance Research Project Agency, Office of the Secretary of Defense; in 1966 he was appointed Assistant Secretary of the Navy, Research and Development; from 1973 he served as Assistant Executive Director in the U.N. Environmental Programme; from 1975 he served as Associate Director of Woods Hole Oceanographic Institute; and from 1977 to 1980 he served as Administrator of NASA.
DeVorkin:Dr. Frosch, we are now to the point in 1977 where you were appointed as the Chief Administrator at NASA. Could you give me a rundown on how and why you got this appointment?
Frosch:Well, I can give you at least the chronology. I don't know if I can tell you why. Frank Press and I worked together, on and off, for a very long time. He was at Lamont Geophysical Observatory when I was at Hudson Labs in the early '50s, we knew each other then. Then, as I have mentioned, when I came into ARPA and began to work on the seismology aspects of nuclear test detection, he was an outside advisor and consultant, and we worked closely together then, and then on the work on earthquake sensing and prediction after the Anchorage earthquake. He also has a house in Falmouth on Cape Cod, and I have a home there, and in fact, was living there because I was working at Woods Hole Oceanographic.
DeVorkin:You were in quite a bit of contact with him through all these years?
Frosch:Yes, on and off. Sometime in the fall after the election, and after the announcement that he was going to be the President's science advisor, he was in Falmouth, and we had a drink with them, and were interested in going to Washington. We concluded that I might be interested in that kind of work, and some one of the Washington jobs in science and technology might be interesting. He wasn't quite sure what he was going to be up to yet, and neither was I.
DeVorkin:Was he aware of your position at Woods Hole?
Frosch:Oh yes, sure, he was then, still is, on the Woods Hole Corporation, so he was active in oceanography.
DeVorkin:I'm harking back to the end of our last session where you indicated that you knew that philosophically your interests in having Woods Hole change were not the interests of the scientists there.
Frosch:Oh. we had talked about some of that, but not in great detail. But at any rate, it was clear to him that I might be interested in coming to Washington, if there was something that made sense. He and I were thinking more about the more obvious task of being administrator of NOAA than anything else. In fact, I was sort of proceeding on the assumption that I would probably have to decide whether I wanted to do that job or not. Meanwhile, when Frank got down to the EOB, he asked me to serve as a consultant to his office, because he was putting together a group to look at technoloqy transfer I think, and he wanted to me to serve on this qroup.
DeVorkin:Was this still in 1976?
Frosch:No, this was early in '77 because he was already there. He wasn't confirmed yet, but he was in the EOB, so it was after the 20th of January.
DeVorkin:Okay. I'm referring to the fact that in April and May you were —
Frosch:That's later. I'm before that. So, here I was a consultant to OSTP. I don't remember the date, but I came down to sit in a meeting in his office to discuss this other task, whatever it was, and I think he stopped me in the outer office before we went into the office, and said, "I want to talk to you for just a minute: and I want to talk to you again afterwards." He said, "'I don't think the NOAA thing is going to go, because the President's team has somebody they want for that, but how about NASA?" I said. "Well, I sort of know about it. You know what I know and don't know about it, why not?" And he said, "Okay, I'll talk to you afterwards." And we talked a little bit, and he said, "Well, let me put your name on the list." He said — and I don't know whether it was then or later — he said, "Well, I think I want to put it on top of the list, and so you just wait and see, and there may be a call from me, saying you're going to come see the President." This must have been February, or the end of February, or early March: I don't remember exactly. I had agreed to go to Geneva and do a consulting task for the UNEP, the Environment Proqramme. So I went off to Geneva, carefully leaving with him and everybody else how to get in touch with me. He knew that I was going, and I knew that I might have to come back and so on. I don't remember the dates exactly, but I guess I flew over on a Sunday, and I was in the meeting on Monday, and one of the secretaries I knew came in terribly impressed because there was a call from the White House. It went through the White House operator, so it was the White House. It was Frank saying, "Look the President wants to talk to you. Can you be back here by Wednesday?"
DeVorkin:And this is Monday.
Frosch:And this is Monday, so I said, "Yes, what time." and so on, and scurried around and got myself on an airplane and came back, and went to see the President on Wednesday.
DeVorkin:This was the first time you had ever met Carter?
Frosch:Yes. And, let me see what I can remember about the discussion: not much. He obviously had read the material he had been given, because he was familiar with what I had and hadn't done, and we talked a little bit about science and management and so on. He did not make a decision then and there. He did not tell me anything.
DeVorkin:Do you recall what his questioning was like?
Frosch:Well, it was clear that one thing that was much on his mind was what to do about the Shuttle.
DeVorkin:What were the options in his mind?
Frosch:"Do I cancel it or not?" was really what was on his mind at that point.
DeVorkin:It's a rather interesting alternative.
Frosch:Yes. What he felt was that he was faced with this very large space project, and he didn't know much about it. Some people apparently had told him it was useless, and it was just a kind of a technology binge, and he simply didn't know what to do about it. I think that he expressed that pretty clearly. That was one thing he would be interested in knowing. There was some talk about the planetary program, and he clearly knew something about that, and a little bit about applications. I was impressed by the fact that he seemed to have some technical grasp of what the questions were.
DeVorkin:To what degree were you familiar with the Shuttle and its progress at that time?
Frosch:I barely knew it existed. I had been in the Pentagon when it was being ginned up by Johnny Foster and Jim Fletcher, and I knew vaguely about it, but it had nothing to do with what I was engaged in at the Pentagon in that period. I just vaguely knew it was there. That was about all. I couldn't have drawn a picture of it.
DeVorkin:There had never been any detailed proposals that had crossed your desk while you were in the Pentagon about the military uses of the Shuttle, or anything?
DeVorkin:Nothing at all.
Frosch:No, nothing at all. That was something DDR and E was worrying about.
DeVorkin:DDR and E.
Frosch:The Director of Defense Research and Engineering. Johnny Foster was worrying about, and I presume, the Air Force was worrying about, but I didn't have anything to do with it. It was just a side show, as far as I was concerned. In the context of the previous discussion, if something was going to happen in space, the Navy would not be in charge of it. and wouldn't really be using it. It might be developing satellites, or uses, but in fact, the launch business was somebody else's problems, essentially. We weren't heavily enaaged in that. I guess I felt it was an interesting interview, and a "You'll hear from us" kind of thing. Within a day or so, Frank called me up and said, "The President's decided, and he would like you to undertake it." I think Frank said, "And he reiterated that one of the things he'd like you to do is report back to him on the Shuttle, what you think we ought to do about the Shuttle." Now, the options ranged from cancel the whole thing, and I don't think he, or Frank, or anybody had said "cancel it," to "what are the consequences and alternatives." It was just a sort of: "do we need this?" Remember, he was looking, as new presidents do, for places to save money to put into his programs. Then there was the question, the perpetual question, "How many orbiters do you build?" There was already some pressure from OMB, which, I guess, had carried forward from the staff of OMB previously, that NASA wanted to build four or five — I don't remember the numbers at that time — and that really no more than three were needed, or maybe you should build one and see what happens, and so on.
Frosch:None of which was based on anybody's having sat down and worked on the problem. It just seemed to be somebody's view who didn't understand why they need four of them. I later characterized that as saying, "Gee, I'm going to run a trucking company, and I really only need one truck, because if I use it all the time, I can somehow manage to do everything," or simpler than that, "I have so many truck loads per day, and one truck can do that, therefore, I only need one truck." Nobody has asked where do the shipments go, and do you have to be in five places at once, and when do you maintain the truck, and so on, a kind of very crude problem analysis.
DeVorkin:Wasn't there originally a whole picture drawn as to how many shuttles there had to be, and how frequently they had to be launched?
Frosch:Oh yes. There was a whole analysis, but it was a very peculiar analysis in many ways, because it was a traffic model analysis that assumed a certain model of what you'd have to carry and worked out the logistics for that. Nobody was quite sure whether that was the right model or the wrong model. There was just a deep suspicion. Now, you have to remember that OMB at this time was Bert Lance, but the guy who was in charge of things that included space was a guy named Elliott Cutler, who is a lawyer, who came from somewhere in the environmental people's defense legal kind of thing. I don't know what group it was, but he is somewhat of a populist and a little bit of an antitechnologist, although also something of a space buff. You keep finding that contradition all the time. But he was all for a kind of Calvin Coolidge solution. "Buy them one shuttle, and let them take turns using it, or get rid of it." I think there was at that point a hope that I would look at it and come back and say, "What the hell do we need it for? Kill it and we'll do something else."
DeVorkin:No one was asking you to go back and look at the relative cost efficiency of expendable launch vehicles as opposed to Shuttle?
Frosch:Well, tbey weren't. Yes, there was some of that, but I would say, initially, the question was not that well formed.
DeVorkin:It was purely a budgetary problem then?
Frosch:Well, it was: what do we do with this program? We've got this big fat program. We are deeply suspicious it isn't necessary. What do you do with it? Now, I looked at some of the cost questions and so on. What I did first was nothing, because I was not administrator of NASA.
DeVorkin:Right. Let's identify this consultant period. You're not up to that yet.
Frosch:No, now I'm up to it. Now I hear I'm going to be administrator, and I start to make my personal plans, but it was clear that the administration wanted me to begin to be there as soon as possible. Jim Fletcher, somewhere in this period, left — I guess around early in May. I don't remember. He had announced he was leaving and he left sometime in that period. Jim was not around when I got there. When I got there he had left. If we met on the 5th, I think it must have been that he came in for that purpose, because my recollection is, when I got there, whenever it was, April or something, he was gone: his office was empty. Al Lovelace was acting administrator, and I was sitting in the consultant's office next to Al, and we were starting to work, but it was clear that the way you did this — I knew this from previous work — that I wasn't going to make any decisions. I wasn't going to say a bloody word, in fact. I certainly didn't tell anybody, including Al, that I had been asked to look at the question "did we need a Shuttle at all.
Frosch:I told Al a little later, I think, and just kept the question in my mind as we went through the briefings. I hadn't even been officially nominated until the end of May, in fact, I was there, and in kind of a half world. You're not telling anybody what's going on until everybody knows, and you are around as a consultant to the administrator who isn't there. People are aware that you are probably going to be the boss, but they are very polite about it. They start briefing you. You know, you run around the agency, and people put on their dog and pony show and so on, and try to convince you that their particular thing is the important thing, and so on. It's a learning period, and getting prepared, and meeting people, and so on — meeting OMB people and the like.
DeVorkin:Was this the first time you worked with Lovelace?
Frosch:Yes: I don't think I'd ever met Al, as a matter of fact.
DeVorkin:Let me backstep a bit.
Frosch:Yes, why don't you go ahead and structure this a little bit.
DeVorkin:Yes, a few questions I just wanted to get out of the way. You indicated that the President had made a decision within a day or two, but do you have any knowledge as to whether Frank Press really made the decision and the President went along, or it was really a real decision?
Frosch:Well, I don't know. All I know is that, as I recall, what Frank told me was that the President had asked for several names, and dossiers, and Frank gave him that: I think some number, like three, maybe four. I don't know who the others were.
DeVorkin:That answers my next question.
Frosch:I never did know. Frank put my name at the top of the list as his recommendation. As I recall, he said. "You know, I made it as a strong recommendation, saying, `this is much the best name, that the next two names could do it, but I'd really sort of put them farther down.'" My recollection is: I don't know whether this is an impression, or I was told, that the President didn't interview anybody else, but I'm not sure. He may have interviewed somebody before he interviewed me. I guess the answer to the question is: I think, what he did was verify Frank's proposal, in a sense, and make it as his own decision. He looked at the papers and agreed that he saw nothing in them to disagree with, and did the interview and decided, "that's all right. That's really what I think happened.
DeVorkin:Okay. Other than the statement about the Shuttle.
Frosch:Well, that stuck out in my mind for obvious reasons as a strong statement. I had had some warning from Frank that, you know, he was likely to ask me what I wanted to do with the agency, what I thought would be important things: and I had done just enough homework so I could talk about the science and the applications, and the technology and so on, and suggest that probably the thing to be sought was some reasonable balance among these objectives, without having a very firmly thought-out program.
DeVorkin:There was no obvious feeling on Carter's part that science should be favored over applications, or anything like that?
Frosch:No. I don't think he knew a hell of a lot about it at this point. He was obviously busy learning most other things. There was one point that I guess was clear in the first discussion: Carter was always pro-Landsat, because apparently, as Governor of Georgia, some of the early Landsat, or maybe even the EROS experiments, were done in Georgia, and they had done something useful for agriculture or something. I'm not sure I ever knew exactly, but he was acquainted with that, and it was a good thing from his point of view. So that was that part of it: but nothing else, other than the obvious Shuttle guestion, sticks out in memory. And it was more than just, "do I cancel it." It was, "If I don't, how many should I buy? What should the program be like?"
DeVorkin:Sure. Okay. I'd like to get to the point where the confirmation hearings begin. Now, during this period you were a consultant, no decision-making activities.
Frosch:Right, and not really existing.
DeVorkin:But you were gathering information on the structure of NASA, how people —-
Frosch:I was getting briefings and meeting people, and what it amounted to was walking the halls and headquarters in a structured way. You know, I'd go down to the AA (Associate Administrator) for applications, and go in his conference room, and he and his people would brief their programs, and so. Then we'd do science, and we did Shuttle; we just walked through the top of the organization diagram, and met with the public affairs guy, and met with the legal guy, and so on, and so on, and so forth. And also, a fair amount of time spent in getting the formalities done. There are all sorts of forms and so on that had to be dredged up.
DeVorkin:You later on effected a reorganization of the organizational structure.
DeVorkin:When did that come to mind? You had a chance to examine it during your consultant period. Did you see any problems with it then?
Frosch:Well, I really didn't think very much about reorganizing the structure much. I did begin to formulate the hypothesis that I wasn't completely happy with the idea that you, in effect, had an organization in which you had the Shuttle, and then you had John Naugle's thing — nothing against John, a first class guy — but then you had a structure in which somehow or other, applications and science and aviation and aeronautics were submerged. That seemed to me to be a little bit distorted in the sense that the Shuttle I always thought of as a means for doing things. Here are the things that you want to do, and it's organizationally a little peculiar to bundle them up so that there's somebody between me and the people who were doing those things, whereas, the means for doing those things are up higher. Now, I can understand why that was, in a budgetary sense, and in terms of where the headaches are. It must have meant that Jim Fletcher did it that way so that he could pay attention to what he thought was his biggest problem, and John Naugle would worry the rest, and he would deal with John. I formulated the view that whether or not that worked, it was the wrong organizational symbol, and that I wanted to have more to do with the output part than I would have in that structure. Left to my own devices, I'm not sure I would have gone to the trouble of reorganizing.
DeVorkin:That's interesting, because you did do the reorganization, so what does that mean?
Frosch:Yes. Well, what it means is that —
DeVorkin:Or, are we getting ahead of ourselves?
Frosch:No, no, what it means is that somewhere in this period, probably in the summer, I guess, after the period we're talking about, I began to get signals and comments from Frank Press and other people that amounted to saying: the way you demonstrate that you are a new administration is you regorganize something. You really ought to do some organization things. That's how you demonstrate you're a manager, which I think is a lot of crap (laugh). Somehow or other, that was impressed on me. I don't think that was Frank's idea. I think that was a message that came from the political types and the OMB types. Well, I did have this piece of the structure I did not like particularly. My feeling for organizational diagram arrangements are second order things. I was more interested in who was working with whom. Now, I had also decided that there were a couple of people involved who were in senior positions whom I didn't think were doing the right thing, and this was particularly true in Applications. The guy's name escapes me now, but somebody decided in the previous period that the problem with Applications wasn't the technology and knowing what to do, it was marketing. They had gotten a marketing guy, and I just didn't like the flavor of it. It sounded as if the technology was sound, but not very well supported, and the whole flavor of the thing didn't sound right for a technological organization, and I just didn't like it.
DeVorkin:Hadn't that long been a problem with NASA, especially, let's say, in the Tiros/Nimbus period, and especially in Landsat?
Frosch:Yes, I guess they had gone through a lot of that, but I just didn't like this view of it as a marketing problem.
Frosch:I have nothing against marketing, if you have a very sound product, and you're figuring out how to market it, how to let people know you have the mouse trap. But what seemed to me was that the flavor of this was that they had it backwards. They were busy dragging people in off the street to see the product which wasn't really yet available. Obyiously, Len Jaffe and others knew what they were doing technically; but I didn't think the guy who was running it had any idea what his product was (laughs), or why there should be such a product, or whether it would do any good for anybody. I didn't like the whole flavor of the thing. I was happy with the science guys, with the science effort.
DeVorkin:Was John Naugle's position changed in this reorganization?
Frosch:Yes; I'll come to that. So, with this pressure, let's do a reorganization. Oh, there was another question that came, not so much from the President, but began to come from OMB and Frank Press, which is important to reorganization. It is: why does NASA have so many centers? Why don't you close a few centers? You know, it's a perpetual guestion. It tended always to focus on Huntsville, largely because the were the engine place, and the mentality of a lot of OMB and political types is a very short-term mentality; and so, they were saying, "Gee, we're almost through with the development of the Shuttle engines. Obviously, you don't need Huntsville. After you finish the engines, you dispose of Huntsville." You can decrease the number of people. And remember, the President came in saying there were too many bureaucrats; you've got to decrease the number of bureaucrats. There was a lot of pressure — "What are you going to close?" In fact, there was a rumor around NASA that the reason I had been selected was because, as I told you, in the Navy job I actually closed something. Okay, so that was mixed up in this whole organizational guestion.
DeVorkin:That rumor wasn't well founded, was it?
Frosch:No, no: as far as I know, it had nothing to do with it. Nobody was thinking about that at all. Oh, there were funny rumors, that since Loyelace and Frosch had both had experience in the Pentagon, the whole place was going to be swallowed up by the Pentagon. In fact, there were people running around at one stage, saying we were brought in to militarize NASA. It was very peculiar, but the only thing you do about these things is you ignore them (laughs), very straightforward. So, we launched, among other things, into "what are we going to do about reorganization?"
DeVorkin:This was after your confirmation.
Frosch:This was after the confirmation hearings. And so we had these two sets of questions: "What is the organization of NASA," and in particular, this subset: "How many centers should there be," and "why don't you organize them in some other way and decrease the number of people," and all of that. Now, somewhere there is a file that has a great pile of paper on this, and the guy you ought to talk to about that is Phil Culbertson.
DeVorkin:How do you spell that?
Frosch:As in the bridge master. He is, I believe, still a special assistant to the administrator. He knows about that, and Dave Williamson knows about that. I think Phil probably has the files on it. What happened was, Alan and I essentially set up a task group which Phil ran to help us figure out what to do. It was going to operate quietly, you know, operate talking publicly in an organization about options for reorganization, none of which you may end up taking. Because they become giant rumors and the whole system halts while people fight what yesterday's reorganization (which you discarded this morning) would have done to them, if you hadn't discarded it, and they don't believe you discarded it. Maybe we overdid that. I don't know. But anyway, we operated in camera with this group. This didn't help things much, because the whole operation was a typical NASA operation: let's get all the data and beat it to death with all possible options and make a big structured thing. I'm making a joke of it. It's not too bad a way to do this, but it was a little bit difficult to look at options for which you needed data from the centers while you're trying to figure out how to do something, and you really don't want to ask the question because it will tell them too much. You don't want to go to the center and say, "Hey, the Administrator asked yesterday when we were reviewing Lewis or Langley. What's the funny thing on the diagram? What do those guys do? And nobody in the room knew; so what is it you guys do? You can't do that. You make irrelevant noisy trouble, you see.
DeVorkin:That does make trouble there.
Frosch:Somebody's got to go and somehow find out what it is they do, so they can come back and tell the study group. We went through — god, I guess it was all summer and into the fall — with people knowing we were doing it, and rumors, and I guess, press questions about it: are you going to carry out a great reorganization? Remember, I had no great stomach for reorganization. The more I looked at the centers, the more I came to the conclusion that, if I were asked today to design NASA, and establish a set of centers to do something, I wouldn't do it the way we have it. It's not designed as a system for what we're doing today. Obviously because it evolved out of a 60-year history, and there it is. The question of whether I think it's a good design is different from saying, "Do I want to go through the awful agony of changing it?" We toyed with just all sorts of arrangements of abolishing centers and moving people, and bundling this and that together, and so on.
DeVorkin:Other than Marshall, what were the other centers considered that were weak?
Frosch:Well, it wasn't a question of whether they were weak. It was a question of whether they were dispensable. That's a different question, whether you could do without them.
DeVorkin:Take JPL, for example?
Frosch:Well, JPL is out of this. JPL is a contract. It is different. JPL isn't in this, but you start out with a question: what centers do we have that are active in aviation? Well, you've got Langley: you've got Lewis: you've got Ames: got Wallops and you've got Dryden. Somebody says: my god, we've got five places doing aviation. They're so small, why do we need Wallops? Do you really need that? Isn't that part of Langley? Maybe we ought to make it part of Langley. We'll have abolished a center, and we can do the accounting at Langley, and all those kinds of things. And I gradually concluded that none of this made any very great sense, that the amount of money or manpower or anything that you would save was miniscule, that you would be damaging yourself for no great purpose. There are questions as to, for example, why do you have Langley and Lewis? Well, the answer is easy. They do entirely different things and have different facilities, and so on. Why do you have Langley and Ames? That's harder. The separation between what they do is partly arbitrary. In fact, it had just been slightly reorganized before I arrived, with a transfer of all the rotary wing aircraft responsibilities from Langley to Ames, and there was still a lot of smarting at Langley over that. There are other things at Ames besides that: there are all those wind tunnels. In the end, part of the decision simply was: it didn't matter whether you liked where the centers were or not: you had a fantastic investment in a mixture of real estate and equipment, and people who knew how to run them. There is no sense in talking about abolishing Ames or moving Ames. I mean, you just have to walk around there to get the point. The guys who say, "Forget about Huntsville," don't realize what test stands are at Huntsville.
DeVorkin:Do you have any names, particularly in OMB or others, as you refer to these people?
Frosch:Well, there's Hugh Loweth, who I think is on the whole a positive guy. He chipped away but he understands these sorts of things, but I don't think Elliot Cutler, for example, ever ended up understanding any of this at all, or was terribly interested. Now, Bert Lance, had he lasted in the administration, I think, would have been a great help in this, because he's a bright guy; and he approached the whole thing without any great prejudice, and was busy learning. You could explain to him, "You can't abolish this because you can't move that." He would absorb that: it would make sense. Some of these other guys wouldn't absorb it: they'd come back three days later and say, "Well. why can't you move it? You know, it's just a big structure." And I'd say, "No, no, I told you: it's got a foundation which is 18 foot thick reinforced concrete. Do you want to rebuild that?" "Oh, no, I guess not," and so on.
DeVorkin:In the case of Wallops.
Frosch:Now, Wallops was an interesting case, yes, because there was something that was important about Wallops, other than that it had a tradition, and there were good people there doing good things. All that might or might not have been important, but the real point is it's unique piece of air space. It is a unique piece of protected air space, and in a sense that meant it didn't matter whether we abolished it from the NASA budget or not: that airfield had to be there for U.S. purposes, and the air space did, and so on. Now, there was a lot of toying around with, "Well, Wallops is small, and can't we get some efficiency by abolishing Wallops as a center and making it a field station of Langley," and so on and so forth. You consolidate this and save that. That all sounds great until you look at the numbers, and you discover you're going to save three people for no particular great virtuous purpose. So you decided it is not worth the destruction of tradition to do it.
DeVorkin:Didn't you decide to strengthen Wallops by sending more of the rocket sonde work down there, or was that later?
Frosch:Well, there was a later thing that had to do with the sounding rocket business which was split between Goddard and Wallops. It seems to me we looked at that at that time, decided to do nothing, and several years later came back to the question. Particularly as there began to be a question of the phasing out of — I've forgotten it and I went to the hundredth launch — of the . . .
Frosch:What's the biggest sounding rocket?
DeVorkin:There's the Aerobee? There's something beyond that?
Frosch:No, the one that's still being used. Well, I've lost it. We can look it up. It's the one that Goddard was running and launching from Wallops, principally: and there was a question of what was going to happen to it, because the Navy was using it. The Navy at some point had decided they didn't want it any more, and the question was: was DOD going to take it over. In the end, we decided really, I think, to transfer it from Goddard to Wallops. That was not a transfer of launch site. That was a transfer of responsibility. That was not a big issue at that point. Then there was the question of what to do with Dryden. Again, it's a little bit like Wallops. Should it be a branch or a field station of Ames, and so on and so forth? I guess, in the end, I decided that those are all very nice theological issues in the theory of organization; but whether you call the guy who runs Dryden the Director of Dryden or the Director of the Dryden Field Station at Ames, is of importance in Civil Service classification, but not terribly important to how the place is run, except from the point of view of morale. The morale question says you may be better having a separate place with its own director, because in fact, the places are several hours apart by car. You're not really going to run the payroll at Dryden any differently when it's this way and so it didn't make a hell of a lot of sense. There is a lot of talk about consolidations and the virtues of scale, and so on and so forth, but in situations like that, it doesn't really amount to a row of pins.
Frosch:So in the end, I guess we messed around with some minor organizational things at centers, and then at the same time had gotten off on this center question, because it kept looming eyery year in the budget. They wanted to cut down NASA manpower, because they wanted to cut down Government manpower, and where the hell were they going to do it. Somehow they could do it with NASA. And of course, the Federal Government has this really peculiar system of budgeting money and people separately. And so, you may or may not get a manpower number that has anything to do with what the budget tells you you have to do. It's very peculiar.
DeVorkin:Well, your budgets in some areas certainly went up. Your manpower went down.
Frosch:Always, always went down.
DeVorkin:And your number of consulting type people went up.
Frosch:Well, that's a separate tale that we need to talk about.
DeVorkin:Yes. Let's go back and talk about the . . .
Frosch:But I want to talk about the reorganization of the headquarters. We haven't really talked about that.
DeVorkin:Right. Let me turn the tape over.
Frosch:What I really wanted to do, having stewed about this and listened to what everybody had to say, what I really wanted to do was to simplify the organization diagram in the following way: (And somewhere this was in one of the drafts Phil Culverton may have.) Make all of the technical portions of equal level. Namely, Shuttle and science applications, aeronautics. One of the things I wanted to do was to have a pure aeronautics bunch, and put the space technology either with the Shuttle, or some place with the space business. There were too many administrative type offices all in a row. External affairs, internal affairs, management, controller, EO, counsel, etc., etc., procurement, and just a whole lot of them. At one stage, I wanted to sort of bundle them together into some kind of administration organization. It became clear, by the way, that you couldn't really do that, for a combination of statutory and regulatory and just plain "public view" reasons. You've got to have access to your counsel. He's got to have direct access so he's a separate office.
Frosch:Your controller, almost statutorily, has to have direct access. You can't do anything about that. You'd better not submerge Legislative Affairs or Congress will feel badly served: and you have so much business to do with Congress that Legislative Affairs had better to be able to run in and out of your office. You can't do it with Equal Opportunity, because everybody had agreed that had to be the highest level. I had already decided that the way I was going to handle that was to say, "I'm the chief EO officer in the agency." Not because I didn't like Harriet Jenkins, who was doing a fine job, but she also agreed that was important to her office if I said that. I thought that was the way to handle it. She was doing it: everybody understood the relationship. I was saying, "Hey, it's on me to get this job done."
Frosch:By the time you were through, you couldn't bundle anything up very much. I guess we bundled a few things together. We arranged some of the internal. administration. I forget what the office is called, the one that Ed Kilgore just retired from.
DeVorkin:What about the associate administrator positions? You didn't change those?
Frosch:No. I guess we may even have created another assistant administrator or so. We did abolish the Associate Administrator, namely, John Naugle's position. That was really because I couldn't see any sense in having the position. I wanted the Associate Administrators for Science, Application, etc., to report to the Administrator. I wanted to pay attention to what they were doing.
DeVorkin:The assistants had been reporting to John Naugle?
Frosch:Yes, the Associate Administrator for science had reported to the Associate Administrator, who reported to the Administrator. And that I wanted to get rid of. On the other hand, I very much did not want to get rid of John Naugle, who I thought was first class. We looked at that problem and decided that, in fact, there was a task that needed to be done, which was what we called the Chief Scientist. The idea was that the Office of Space Science was science in the sense of the science objectives in space, but it was not science in the sense of the science that underlies applications, or science in the sense of the science that underlies aeronautics or whatever. There was this across the board task of councilor for science, so to speak. Plus, it was clear we needed to do more than we had been doing with regard to relations with universities. There was also the question of advisory committees and councils, which we had to reorganize, because Carter came in with a thing about consultants, which meant everybody should have fewer of them. When the President says that, nobody will listen to reason any more, so you have to do something about it. We had to reorganize the advisory committee system which had its difficulties, anyway. We needed somebody to watch over that, and I convinced John that this was a package that would interest him.
DeVorkin:How much convincing did he take?
Frosch:Well, it took a fair amount. It was not that he was opposed to what I was doing, it was just that he was saying, "I was thinking of retiring about now, anyway. If I'm not going to do that and if I don't quite see that there is a useful job for me in the agency, I'll go do something else." It was a question of saying, "I can't tell you, John, do this, and it's obviously a big job, but there is a package of problems that I need help on." He participated in structuring the job. We sat together and worried about it. Also, somewhere in this period, I guess, we were reorganizing the advisory committees structure and creating the NASA council. In part, that was sleight of hand.
DeVorkin:What do you mean?
Frosch:Decreasing the number of committees in the OMB regulatory sense is what we were doing. Previously, each of the AAs, had his own advisory committee. By inventing the structure of a council in which the advisory committees were subcommittees, the change counted as a total decrease in the number of advisory committees in the official system. It's the kind of thing that makes people over in the White House, who have to count numbers politically, very happy, because they can announce: NASA has abolished five committees. It's all Mickey Mouse, but the trouble is that people come in with ideas on what they ought to do. They made political speeches about them before they understood the subject. They make a few more political speeches before they bother to learn. Then when they discover that what they have said really doesn't quite make sense, everybody has to rush around and produce enough cosmetics so that they can say, "Look, we did it," without actually damaging the system. Fortunately, it mostly happens in small things, like the number of committees, and so forth.
DeVorkin:Yes. One other center that we hadn't mentioned did come up on your daily sheet here: and on September 16th you had a meeting with Jastrow and Naugle.
Frosch:Well, that was GISS. Yes.
DeVorkin:GISS, Goddard Institute of Space Science.
Frosch:Yes, now, GISS is a funny little thing that was really originally built around Jastrow. As you may know, it's a field station of Goddard at Columbia University. Partly directly Civil Service, and partly contract. It is a very nice little scientific place, doing some very good IR astronomy and some excellent weather modeling and so on. The problem was that Goddard wanted to swallow this up or abolish it because they didn't feel they quite had their hands on it. It was complicated by the fact that Jastrow was talking about retirement, and some of the others, and he, were worried that if he retired, then the whole thing would go pop. We were discussing what to do with this, particularly vis-a-vis the university, and in case Bob Jastrow did retire, what would happen. That's what that meeting was about.
DeVorkin:Obviously, nothing happened.
Frosch:Obviously, nothing happened. It's another one of these cases where we could tidy up the diagram by abolishing it, but you don't think you are going to abolish the work, and the people are good, so what do you want to fool around with it for?
DeVorkin:Yes. What was the character of university relations as you came in, because I know that there was a very big push in the Webb administration for major university contacts as there always had been, but that that had been on the wane.
Frosch:Yes, and my recollection is that I had a sense that that was very uneasy, that the universities weren't quite sure what we were going to do and weren't happy. So one of the tasks that John Naugle took on was ambassador to the universities. He specifically, at one point, got the task of finding out what's bothering them, and whether we should do anything about it. I don't know that we ever did anything structurally or definite about it, but John was able to improve a lot of relationships. I have a vague idea that there were recommendations on that subject which I accepted, but I don't remember what they were, or whether we did anything. There would be a memorandum somewhere that describes that.
DeVorkin:Okay. Getting back to the applications part where there was this person.
Frosch:What was his name, Johnson?
DeVorkin:We didn't identify him. Was it Brad Johnson?.
DeVorkin:. . . who was more of a marketing person. How did you resolve that problem? Did you do it by reorganization? Or by putting a different person in there?
Frosch:No, he shortly thereafter resigned.
Frosch:I didn't fire him. I didn't have to fire him. By the time I was making up my mind that I might, he got the idea and decided that the way I was talking and moving, he wasn't going to be very comfortable with what we were doing, so he left. I suppose I would have been forced to fire him or ask him to leave.
DeVorkin:So that by the end of October then, you had abolished the office of the Associate Administrator and created the chief scientist, and all of the assistant administrators reported to you.
Frosch:Now, the next complication was the centers in all of this. Now, the centers had been in the situation that they were essentially reporting to Naugle, but in effect, more or less reporting to the AAs: and this was very unneat, in the sense that there is not a very good isomorphism between what the centers do and what the AAs do. It just doesn't fit very well. That seemed to have the result that there were things that went on in a particular center that were important to NASA, but not important to the AA with whom they had to deal.
Frosch:That didn't work. There is also the complication that the AA to whom the center director reports for program can't really do anything about the administrative problems of the center, which have to be done by the administrative and managerial people. In an attempt to deal with this question, Jim Fletcher had set up what amounted to an office for center administration, which was another AA. Here again, my name memory is terrible. We can dredge it up.
DeVorkin:I have an organizational chart in my office. Would you want me to get it?
Frosch:Of which incarnation?
DeVorkin:Well, it has you up at the top.
Frosch:Well, okay, but if it's not before that, it may not have the right people.
DeVorkin:You mean, before the reorganization. No, it isn't. I'm sure of that.
Frosch:Well, anyway, we can dig this guy's name up. I can visualize him. I know him; he's a yachtsman; he's very good. He wanted to retire, anyway, and so it didn't bother him that we wanted to rearrange things. In fact, he had said he wanted to retire. He was waiting around through the transition, I think. At one stage we looked to see whether there was some way to rearrange the work among the centers and the organizational headquarters at the same time, so that the isomorphism would be pretty good. That turned out to be impossible. Things were just in the wrong places. You had the right wind tunnel in the wrong center for that purpose, and so on. We ended up saying, "Look, it's probably a good thing for the centers, anyway, if they get a feeling of not being so submerged in the organization." We changed the organization diagram so the center directors reported to the administrator.
DeVorkin:Now, you clearly are creating a much bigger job for yourself.
Frosch:Right. That's the immediate criticism. You've got all these guys reporting to you. How can you run the place when you have all these guys reporting to you? My reaction was: that's the reaction of somebody who reads the organization diagram literally. What I'm going to tell the center directors is: you report to and have access to the administrator. Don't come in the office every day! You really know who you have to do programmatic business with. This is the various AAs for various things, and you know who you have to do administrative business with, the procurement office and the legal office, and the comptroller's office, and all the rest of it. Do your business. Don't come into my office and put me in the position of telling you: you've got to go to talk to Noel Hinners about that, because it's a program you're working with Noel. I want you to know, we're going to have center director meetings a couple of times a year, and we're all going to be together, all the AAs and so on. You're always free to call me or come see me, but I expect you to do all the work yourself. I don't want you running to me (and I told this to the AAs, too) because you've got an argument with Noel, unless it's the kind of thing where the two of you come marching in and tell me you can't solve it, and I've got to. Or Alan will do this, whatever. Now, one other thing which is important is that Alan and I early came to continue a traditional arrangement. He would be, in effect, deputy administrator and general manager. The deputy administrator would attend principally to the running of the house, and that I would do those things that only the administrator could do. Like see the President, testify, and so on, make those decisions that only an administrator can make, and worry about the philosophy and sense of direction of the agency. This was a conscious discussion which we had almost immediately after Al offered to resign on principle because I was a new administrator and might want to choose somebody else, which I rejected. He was basically running the agency. Now, there's another reason for doing this which is important. One of the things an agency would like to do is keep its business, insofar as is legally possible, within its own hands. That is, the administrator doesn't want to be in the position of referring decisions that are properly his to the White House, which is the only higher authority. Now, that meant that for certain classes for decision, the decision has to be pushed down one level, so that the administrator is not a part to the decision, but can be the court of appeal. Otherwise, you've got to go out of the agency for appeals and that's not healthy.
DeVorkin:Was this new in the policy of NASA, or were you continuing this?
Frosch:Well, it beats the hell out of me. I don't know.
DeVorkin:Okay, but you made a conscious decision to do this.
Frosch:We made a conscious decision to do this. In fact, Alan and I worked it out. For example, jumping ahead, when we came to the SES period and all the business of evaluating individuals, in spite of the fact that it was a terrific load, we consciously made the decision that Al would write the fitness reports of all the other senior executives, and every senior executive knew that. The only official in the agency whose fitness report I would write other than my secretary, aides, and so on, was Al Lovelace's. If an executive had a guarrel with Alan's request on them, they could go to the administrator: he had not been party to the decision or the quarrel, but could then, within the agency, adjudicate it. It didn't happen, but I think one of the reasons it didn't happen was because, in a certain sense, nobody got terribly aggrieved. They knew what was going on. We did that consciously, and that threw more workload on Al. But consciously he wanted it, and that left me with more time to worry about: what's the applications business for; why are we going to Jupiter; and that sort of thing. That's why I was less concerned about: gee, you have 27 people reporting to you. The answer was, there were 27 people who have the right to come to me, and if somebody says to them, who do you work for, they work for the administrator; but I don't expect big crowds in my anteroom. In fact I didn't have that difficulty with directors at all. There must have been a few cases where a center director would call and come to see me. And I'd say, "What are you bringing this to me for? Why don't you go do this and so on?" Generally, they did stay out of the office, but they did feel free to call me. I would get a phone call from a center director saying, "I thought you would like to know that," or "I am going to do this," or "I think there's a terrible problem coming and you ought to worry about it," etc.. Sometimes, I'm happy to say, they would call up and say, "Hey, somebody made a great discovery. I thought you would like to know."
DeVorkin:Did that happen?
Frosch:Oh yes. Well to jump ahead, the most interesting case was not a center director, but it was interesting. Now, maybe this wasn't new. Maybe it's always been that way, but it was interesting that when the guys in Voyager 1 found the volcanism on Io, Brad Smith called me directly to tell me that he was about to announce this. I remember it very well, because the circumstances were beautiful. The Voyager encounter was happening, and I was in the staff meeting. All the staff were in the conference room, and the secretary comes in and says, "Brad Smith, who is the visual images coordinator, would like to speak to you, and he says it's urgent.What new catastrophe, I thought? The little green men are talking to us. They have illuminated signs on Jupiter! Anyway, so I go and Brad says. "I want to tell you that I am going to make the following announcement at a press conference tomorrow: that there are live volcanoes on Io." And I said, "Brad, I don't want to question what you're doing, but I do know there have been some cases where on the basis of initial data we have made startling announcements and have had to retract them. Are we being careful enough?" He said, "Look, we're sure of this. We're careful enough." I said, "Okay, you guys have the data. I haven't got the data. I don't want to stop you: go do it." I went back into the meeting, and I said, "Guys, you ain't going to believe me, but ... "(laughs). I felt I had some reasonable rapport with NASA. One other thing I did which is not organizational, but was an automatic continuation of the way it had been in the Navy, was to accept many occasions to go to the centers. In fact, Al and I constructed a change in the NASA awards system after the first awards we gave in the fall of '77. The first awards ceremony was really a monster rally, because the custom was that all NASA headquarters organizational level awards, all sorts of all NASA awards, all the NASA medals (the ones that were given by the administrator as opposed to the ones that could be given by the center directors), were given in one glorious two day award-giving orgy at headquarters. We discussed this, and decided that the trouble with this system was that we bring the families, we couldn't bring the colleagues. Wouldn't it be nicer to give the award at the center like the British motto: "Justice must not only be done, it must manifestly be seen to be done." So the guys, colleagues and families could say, "Joe got an award and the administrator was here to do it, and we all went and saw the ceremony."
DeVorkin:Right, and they got to see the centers, too.
Frosch:Yes, that was the other part. The counter argument was: "but it doesn't have the prestige of calling the man or woman to Washington for a medal." Well, we looked at it and we decided we're going to try it the other way. And I guess we did it either once or twice this way. This last year, last fall, we didn't do it this way because of the Shuttle launch preparations. We just couldn't find the time to do it. Now, how the hell are we going to do this? We have whatever, 11 centers, is it, 10 centers? It's just impossible. You're not going to do anything else all fall, except go to centers. What we decided was we would arbitrarily split them, by how our schedules were. I'd do half the centers and Al would do half the centers in year one; and then in year two we would try to swap over, so if I hadn't been to the center last year, I'd be there. In addition to that, I took to accepting, and even slightly encouraging, invitations (if I could fit them into the schedule) to come to centers for some event, whatever it was. Or just to go to, for example, Johnston, and plough around and see what they were doing. Any reasonable excuse that could be fitted into the schedule. I discovered an interesting thing, that, after awhile I would get to a center, I guess it happened at Ames first, and Sy Syvertson, the second or the third time, I remember, stood up and said....
DeVorkin:Who's that? Cy?
Frosch:Syvertson, who is now the director at Ames, was then, I guess, the deputy under Hans Mark. It was either while he was acting or was director after Hans came to Washington. He stood up and said, "I just want to remind eyerybody that Bob Frosch has been to Ames more often in his first year, or two years, or whatever it is, in office than any previous administrator has ever been to Ames." You know, I thought that was a good thing. And we'd fit as much time in the schedule as we could, preferably a day and a half, so there were, two, three, four-hour sessions to go look at things in the labs, and some kind of a dinner or a luncheon. What I wanted to do always was somehow to have an informal luncheon with the senior directorate of the place I accommodated —whatever local social custom the local director wanted. But it was clear that I was available to them, to show me the place, have me talk with people, have a press conference, make any other announcements they wanted to have made by the administrator, whatever was reasonable. It varied from center to center. It was interesting. Goddard did an interesting thing one time. They, in addition to whatever else I did, put me on their internal TV set, with a statement which could go out to other centers. I saw more of the technical guts of centers than I would have otherwise seen. I got actually to see a piece of the work, which I found very useful because it (partly it's a little Pooh-Bah, you remember Pooh-Bah in the Mikado. He was asked why he had said something. He said, "Well, to add verisimilitude to an otherwise bald and unconvincing narrative."), well, it gives a nice sense of realism. It also is useful for my understanding, but it would give a nice sense of realism when I could say to an audience, or say to a Congressional committee, "I saw an interesting piece of technology. It was a very interesting experiment going on and it looks like this." I got a better feel of what the hell goes on in the organization.
DeVorkin:Were you actually able to put that to use on the Hill during that period?
Frosch:Oh yes: oh yes. Yes, because for example, when somebody would say, "What do you do with Landsat?" I was able to say, "Well, the census bureau was doing the following." I could describe it much better than if I had seen a paragraph on it, because I had stood behind the guy at the console at Goddard and had seen him play this game of reproducing what the census bureau was doing. That way I had a sampling feel for what's good and bad in a center. I guess I came out of the Navy priding myself with the feeling that I could walk through a laboratory for an hour or two and smell whether I thought it was good or bad in terms of what I was shown. Who it is they have set up to do the briefing? What happens in response to a question? NASA's particularly good in this regard — the NASA habit throughout the agency, and I can't think of any exceptions, is when you go around the agency, or when there are briefings at headquarters, in distinction to some other organizations, the briefing is likely to be done by the lowest level person who knows how to do it, who is doing the work. I always had the feeling — there can always be a certain amount of Potemkin village in this, but still, I always had the feeling that yes, it was set up, and they mopped the lab because the administrator is coming, and all the rest of the VIP baloney, but still I was listening to what the project was about from the junior engineer, who was set up because he got to tell the administrator about it.
DeVorkin:All set up?
Frosch:I began to realize the altitude difference from the point of view of the junior engineer. He was, I mean, set up in the sense of being bucked up, of being happy and proud, because he's going to tell the administrator, who after all, is somebody who talks to the President occasionally, what it is he's doing.
Frosch:This is a good thing for morale. I always feel I have to say, and I think it's right, that at this point, you don't think of the administrator as being yourself quite. It's like a person that you're carrying around. You have to separate yourself from the person or otherwise you'd get very funny about who you think you are. Furthermore, I was able to get a flavor. If I ask a question, do I get an answer? Is this guy reading from the script he memorized last night? Is this really the guy who knows? You can very rapidly get an idea of whether a place is very good, or not very good, or how much depth they have. No other way to do that. I think it's useful for the administrator.
DeVorkin:Did every center behave this way?
Frosch:Generally. There was one funny incident. Huntsville, I think, showed me exactly the same demonstrations, or about half of the same demonstrations twice on two visits, and I filed that in my memory, too, you know (laughs). It meant two things. It meant that whoever was supposed to be tidy about such things wasn't as careful as he might have been. It also meant they didn't have all that much depth of stuff that they thought was good demonstration stuff. Now, in part, that's right, because they didn't at that stage. There was one area at Huntsville where they were doing some materials research where I was very polite, but I didn't think a hell of a lot of what they were doing. Other areas where they were doing some metallurgical stuff where I was very impressed. And of course, their rocket engines work, their engine testing and their understanding of that is superb, and that's always impressive.
DeVorkin:They were building a space telescope down there.
Frosch:Well, they were responsible for space telescope. It was being built by contractors.
Frosch:I was very puzzled by that. My first reaction was, well, somebody decided they've got to put something in Huntsville just to anchor it, to make sure somebody doesn't abolish it because there's nothing going on there. I think there was such a motivation.
DeVorkin:Yes, C.R. O'Dell. O'Dell has been down there for over a decade, hasn't he?
Frosch:Yes. On the other hand, it turns out they have several people, I guess it's O'Dell, and some others, so they are very good on what I call optical engineering. There is no guestion about it, and they have some scientific depth in it; and it was a perfectly reasonable place to put it. They do understand that.
DeVorkin:He was due on that.
Frosch:They do understand some of that.
DeVorkin:He was there because of the space telescope. He was sent there, I thought.
Frosch:Well, I think that's right, but whatever the original reason was, they now have that kind of a capability; and the proof of the pudding will be in the eating, but I think they've done a good job on it.
DeVorkin:Okay, let's go back and talk about your confirmation hearings themselves.
Frosch:Ah, yes. Okay, I've gone forward.
DeVorkin:That's okay, because we've talked now about the restructuring.
Frosch:Sidelights on the hearings. As a matter of courtesy, one normally goes to one's senators and meets them, and asks them whether they would, so to speak, not quite sponsor you, but would like to stand up for you, since you are from their state.
DeVorkin:You went to the ones from Massachusetts?
Frosch:So I went — how did this go? I went to Senator Kennedy, and that appointment was arranged. Everybody assumed that he would not have any interest in this at all, but would politely meet me and say, "Well, unfortunately, I can't go over." He right off volunteered to come over to the hearing. And it seems to me we went to Mac Mathias.
DeVorkin:This is not on here.
Frosch:You ought to find that. It will be in the confirmation hearing thing.
Frosch:It was not the other Massachusetts senator. I think it was Mac Mathias, and the reason for that was: he was the Republican who was the closest to us. That was a NASA idea. The excuse was that all of the time that I had been in Washington before, I had lived in Bethesda, so he was sort of my senator. This is my recollection, and I may be fabricating that. Kennedy I am sure of. The Mathias thing is a little fuzzy, but he came, too. That you can reconstruct because it is all there.
Frosch:And they both said their words, which was very nice, and unusually strong support, I would say, to have two senators. The hearing itself was funny. If you haven't read that, you ought to read that hearing. I was amused by it. Of course, it was all the usual stuff about previous experience and so on, except for Jack Schmitt of course, who had his own ideas about it. The main thing I remember with Jack was his saying, "Well, you're supposed to understand these things. Suppose this Congress authorized and appropriated whatever the right amount of money would be — and I won't really ask you about that —
Frosch:A manned mission to Mars. What's the earliest you think NASA could carry out the manned mission to Mars?"
Frosch:I said something like, "Well, I've never really thought my way through that problem; and I don't know an awful lot about it, but I guess I can try to construct a solution for you. He said something like, "Well, if you're going to do this job, I presume you're capable of it. Go ahead and construct the solution."
Frosch:Right there. So.
DeVorkin:You hadn't been prepared for this or anything?
Frosch:No, I thought about it and began to calculate. I can't remember the details. That you can read. I worked my way through: We'd start now, and we'd have to prepare thus and so, and so on. And I ended up saying, "Well, here it is 1977. My guess is, along about ten or twelve years from now, about 1990-ish, we would be ready to launch. And now, Senator, I have a problem. I don't remember the round trip time from Mars, I would guess it would be a couple of years, but I don't know what the positions of the planets are." And he said, "Well, it would be 18 months round trip, not counting the time on Mars," (laughs) and we went through that and I ended with with, you know, some half-assed construction of a mission model. I think he was having fun, as Jack is wont to do, but also, it's a reasonable thing to ask, for the reason he said. You know, if you're supposed to be so good at this stuff, show me how you'd approach the problem (laughs). It was clear it was not a crucial question: if you don't answer the question right, I won't vote for you. It's also clear by the way, that at that point for an appointment like that, the way had been prepared, and somebody had talked to all the members of the committee and checked around. There wasn't anybody going to blackball me or anything like that. In that sense you know it's a pro forma hearing, unless you trip and fall on your face in some stupid way, because they've already discovered that there's nobody who finds you offensive or has a political problem, or whatever.
DeVorkin:There are people like Don Fugua, who is definitely your friend.
Frosch:Well, but he's not in the confirmation hearing.
DeVorkin:Oh, I see, I'm sorry, I thought —
Frosch:It's Senate. It's advice and consent of the Senate.
DeVorkin:Oh, okay, that's right.
Frosch:Meanwhile, you go and make your manners on the House side.
DeVorkin:You met with the Congressmen after the confirmation hearing.
Frosch:After the confirmation hearings, yes. You see, you don't go to them beforehand, because there is a vague feeling that it will offend the Senators if you're polite to the House before Senate hearings. I don't think there's much to it. They're all big boys, but still, there's a certain etiquette to the thing. As soon as you're there, then you go see the people you're going to have to deal with.
DeVorkin:So the confirmation hearings were somewhat pro forma. They were friendly. What were their concerns, though? Did they bring up the Shuttle?
Frosch:They were worried. They brought up Shuttle; they brought up goals for NASA.
Frosch:: Long-range goals for NASA. They brought up planning. I had been warned that they were going to bring up planning. You know — what do you think about planning? And there, I have to admit, I slightly shaded my real view. My real feeling about planning is that plans are worthless, and the process of planning is very useful. I shaded it enough that I didn't bother to say that plans are useless because they are obsolete the day they are made. The planning process is useful because as soon as you discover they are obsolete, you have done all that work and you're capable of making a new one, in terms of the new circumstances.
DeVorkin:Was this Harrison Schmitt's question?
Frosch:I think that was Jack Schmitt, yes, who was asking that. That's easy to check, because it's all in there.
DeVorkin:Right. What do you feel, beyond the hearing transcript and that sort of thing, what did he want? What was he asking you do do? I mean, did this Mars mission ... It seems to be, that's certainly a long-range major goal-oriented thing.
Frosch:That what he thinks is the next important manned large scale thing to do. He would like to make that a goal, I think — a manned exploration of Mars. It was on his mind, and it was a good kind of test question to ask. Let me make one other comment which I think is important for the record. There was a question asked at the hearing — I don't remember by whom — to the effect of: How have you been treated as you come into NASA? You remember, at the beginning of every administration there is a deep suspicion of the bureaucracy.
Frosch:"The bastards are fighting." those kinds of things. Somebody asks the question: how had I been treated?
DeVorkin:By your own people?
Frosch:By my own people, the NASA people who were briefing me. I said then, and it's true even more, looking back, that the agency wanted an administrator. And if the guy, if the person, were reasonably presentable and made sense, they were going to fall in step behind this person, and really take the attitude: "you're going to save us." I was really welcomed very thoroughly. Whatever muttering in the corridors there may have been about: "well, you know, he's going to militarize us, or this or that," there was nothing but full support. In fact, the flavor of the briefings before the confirmation hearing, with the counsel saying, "You've got to worry about this and that, and conflict of interests and you're clean, but there's this and that kind of question that might be asked, and you ought to say thus and so." The f lavor of it was "You have got to do well. You're our guy, see, and we're going to get you in the position to be the most impressive confirmation hearing there's ever been because that will impress people that NASA's in good hands." If they're impressed by the new administrator, then that's good for the agency.
DeVorkin:That's in their own best interest.
Frosch:Yes, but apparently it doesn't happen in all agencies, I gather from my reading. Though it was true in the Navy situation, too. I didn't have any trouble there. Now, maybe it's because I don't come in with any great suspicion that somebody's going to do me in, or that I have to fight the bureaucracy.
DeVorkin:Was there any question as you came in, in the first few introductory months while you were being briefed, that you may have had some strong mission that was different than NASA's mission? Did you perceive NASA's mission during this consultancy period? Did you see what their mission was? Was it clear to you? Or did you see that you had to create a mission? I mean, the problem with the Shuttle.
Frosch:No, I didn't approach it that way. I guess, from that point of view, I take a rather passive approach. I'm not sitting there consciously saying, "What's the mission?" I came in and said, "Well, the first thing I'd better do is learn as much about this place as I can, and absorb and absorb and absorb. Just make comments, you know, after it's reasonable for me to make comments, on subjects as they arise. If I want to change things, I begin to engage in a kind of plastic deformation of the thing, rather than big change. I don't like the business of: "let's reorganize eyerything and change eyerything, and put it on a proper basis," and so on and so forth. It seems inconceivable to me that somebody who hasn't been in a place like NASA, and doesn't know what is going on there, can come in and feel that within six months or a year they already know so much that they can redo eyerything that's been done for four years or eight years or ten years. My inclination is to torgue the thing around slowly. I was not even happy about the fact that we had to consciously do an all-at-once reorganization. Now it's true, you have to do that; but you have to do that because the Civil Service system by design is so brittle that you can't reorganize from time to time in a casual way the way you could something else. It's a big paroxysm, I discovered, of everybody rewriting their job descriptions and so on. Whereas, I would have preferred to tinker with the situation a little bit at a time. You know — I don't like that little piece there. Why don't we try it over here? — that kind of thing. There's no simple way. Eyerybody takes it so seriously that you can't do it more than once, and it isn't good for the organization.
DeVorkin:Yes, but you knew that Jack Schmitt was looking for goals, looking for missions.
Frosch:Well, I didn't know much about that at that point. It became more and more clear what he wanted. Now, what happened to me on the goals thing was that I developed a different theory of NASA.
Frosch:Than the grand and glorious single goal. In the first place, I was puzzled because people kept saying, "NASA doesn't have any goals." And I said to myself, "Here's an organization in which when you say: what are you doing, you've got the science guys saying, 'We're exploring the solar system by going there, and we're exploring the universe by looking at it.'" We have an articulated program to look at the solar system which had been developed with the Space Science Board over the years. We have an articulated program to explore the universe in all the kinds of radiation there are, in a systematic way. That's not having a goal? That's a goal. We've got an applications group which, once it got sorted out, was saying, "Now, we're systematically finding out how to look at the earth in a synoptic way. We've got a program for understanding the chemistry of the upper atmosphere. We're building a thing called SEASAT because we want to find out what we can do for the oceanographers. WWe don't have, but would like to have, the next generation of communications technology. All of which makes perfectly good sense. We've got the manned people who are saying, "What we've got to do is get this Shuttle going, and we've got all sorts of lovely ideas beyond that, but we've got to get this Shuttle going."
DeVorkin:Is that where the goals became cloudy, in the manned part of it?
Frosch:Well, now let me come to what I think the problem was. The problem was that NASA had a very serious post-Apollo hangover. That is, a bad case of: "Why don't they love us? Why isn't the whole country on the 50-yard line cheering and yelling, GO, GO?" Here you are saying there is this well-structured program in which we are doing science, and we're doing applications, and we're building the Shuttle, and will build further on it. I had a gradually articulated this attitude: "You build on the Shuttle, and so on." "Where are your goals?" My reaction was (I'm telling you the result of several years of mulling this over) that Apollo contained the seeds of its own destruction. If you say, "Our task is to put a man on the moon and get him back safely." the answer afterward must be, "Okay, fellows, you did that, and then you did it again, and then you did it again, and you did it again. It was scientifically very interesting and the engineering was very interesting." But it all looked the same to the public, who said, "You went to the moon!" So you killed yourself. The project was self-terminating. You then had a big paroxysm of running around saying, "What are we going to do for an encore?" Nobody was sure. Meanwhile, all of this baggage that got carried along with Apollo turned out to be the real show. You know, you did the Mariners and so on, because you needed them, or —
DeVorkin:Oh, the Rangers and the —
Frosch:The Rangers, because you needed them, and then you did the Mariners and some of the others to give the scientists something to do so they would stay out of your hair in Apollo. I'm reconstructing what I've been told. It turns out that all this business of comparative planetology is really right, you know. There's this payoff, and you've got a line of productive work. Meanwhile, it turns out, all the stuff you were doing because Jim Webb wanted to find a way to be useful, really is useful. You made a communications revolution twice, and it is still going on. And there's other technology. Landsat is beginning to be used, and so on. What are you talking about goals for? You're in business with accomplishments and ongoing tasks.
DeVorkin:You're talking to Schmitt at this point?
Frosch:Well, I never really talked to Jack quite that way, but I said, "I don't quite understand what you want." I can see one virtue to articulating a goal; that is, you put up a flag. You put up a banner or a torch, or whatever, and so on; but it is hard to do that when you've got so much unfinished business. I didn't think Jimmy Carter was the kind of President who was going to do that. I was not very excited about the Goals for Space exercise. We were shoved into that.
Frosch:Well, the business of writing the Carter Administration goals for space.
DeVorkin:Yes. Where did Frank Press come into this? Did he side with Schmitt.
Frosch:No, no. He was another force and ...
DeVorkin:It was a three-way.
Frosch:Yes. I'm not sure I remember the genesis of the details, but we found ourselyes suddenly in what had become a slight political tug of war over who was going to control writing space policy.
Frosch:Space policy. That's what it's called.
Frosch:Dave Williamson is the guy who has the best memory on that, because he did a lot of it in detail. The administrator kind of comes in and out and goes to the major meeting, but meanwhile, there are other guys on the staff who are struggling with the White House staff and the other agencies and the DOD, and wordsmithing and compromising and arguing and bringing me things. And I'm saying, "No, no, no, I won't do that," or, "Well, yes, it looks all right to me. Whats wrong with it?" Or sometimes they are saying: "Look at this atrocity," and I'm saying, "Well, it doesn't look so bad, you know (laughs)." So all of that is going on. Then, what happened in the end — one thing you ought to do is take the actual policy, the unclassified part (there's one classified part, but that won't affect any of this) — because it was kind of an omnibus; it was more than just NASA. It was also DOD policy.
Frosch:Take that and sit it down next to the press release on the policy, and you'll discover the policy is, on the whole, better than the press release. A couple of people tinkered with the press release at the last minute and added some more water to the stew.
DeVorkin:This is NASA people.
DeVorkin:Oh, I see.
Frosch:This is over in Frank Press's staff, not Frank, but some of his staff people. Silly things. I think it must have been Elliot Cutler who put a totally unnecessary comment in the policy, like, "Of course, we always will have the flexibility to reduce budgets, if necessary." You know, that's a dumb thing to put in a policy. It didn't add anything. Eyerybody in the world knows you have the flexibility to reduce budgets. What that did was take a policy that was basically sound (it wasn't a flaming torch policy, but it was a policy that fitted with what we were really going to do in the next four years) and made it look like less policy than it was. You could have said what I said in a ringing way. Say, "By god, that's a really forward-looking policy. They really want to explore the solar system within the available resources." I have no objection to that. They really want to make applications. They really want to complete the Shuttle and build on it and make it useful and so on an so forth. And everybody would have said, "Oh. that's great." and Jack would have complained, "There are no forward-looking 30 year goals" and something might have been done about that. At this point, it's early in the administration, remember, this is still '77. And they all clutched up about what other people would say about it. One of the problems that democracy has is that, by design, it's government by amateurs, and periodically you get a new crowd of amateurs in, and it takes them a while to learn. They come in and discover that gee, from the inside looking out, it ain't the way they thought it was going to be when they were on the outside looking in, and they mill around a lot. They get better at it after awhile. A lot of these things that are done early in an administration are not so good.
DeVorkin:In terms of such major NASA commitments, such as Shuttle, what was the effect of this space policy at the time, or was there any?
Frosch:(whispers) It didn't have any effect at all.
Frosch:No effect. No, it was a statement of what was known to be true. It didn't affect anything. See, what Jack Schmitt wanted was something that would affect something. He wanted the President to say, "We are going to, over the next 50 years, explore the solar system, manned exploration of the solar system. That's what we're building up to." Because, then he can use that to say: "To push on, the next thing we need is a manned space station," and if somebody says, "What do we need that for?" he answers "Well, the President already said we are going to explore the solar system, and therefore this is just an initial step on the way." If you don't have that, you say, "We're going to build a space station," and everybody says, "What are you going to do that for?" And so you have to define it and justify it in its own terms, whereas in fact, you're not quite building the Shuttle just to carry satellites up and back. You're building it because it is the next step in learning how to do engineering for whatever you want to do thereafter — which you know damn well probably is some funny combination of robot and human work on the moon eventually, and work on Mars, and so on and so forth. But you haven't ever articulated it. Now on the other hand, a long-range space policy and 69 cents will get you a cup of coffee. It doesn't quite do something. Okay, so President X enunciated a space policy, and four years or eight years later, President Y who may or may not be the same party, doesn't give a damn. Besides, it's a policy. "Well, okay, so the U.S. was going to do that." and you can change your mind. Lyndon Johnson, if he had not gotten himself identified with the space thing as Vice President, might well have said, "Well, the hell with Jack Kennedy's funny idea about going to the moon. Let's say we're going to do it in 20 years and not in ten years. I need the money for the Great Society."
DeVorkin:Well, in that case, Johnson was ...
Frosch:He was the heir of Kennedy, so perhaps he couldn't do things differently.
DeVorkin:He was also the person who was really pushing it.
Frosch:He really pushed it, in fact; but in principle, that could have happened. He could have, so to speak, changed his mind as president, and say, "That was a good game for vice President, but now I want the Great Society, and never mind these details." Well, he wasn't that kind of guy. Suppose that transition had been three years earlier, from Democrat to Republican, then the Republicans might, on principle, have wanted to label that as a giant liberal boondoggle, or something, however they wanted to label it. There's a certain amount of political phasing that's important.
DeVorkin:But even at a deeper level, and something that I certainly want to have your thoughts on, is the whole efficacy of manned space flight, and all the rationales that are built up for it ...
Frosch:I could be happy to go into that. Do you want to do that, or do you want to clean up the Shuttle-President question a little bit.
DeVorkin:Well, I think that would lead into it nicely if we finished up with the Shuttle.
Frosch:Okay, let's clean up the Shuttle a little bit. Okay, so here I was faced with "What the hell do I do about the Shuttle?" My first problem was "What the hell is a Shuttle?" in more concrete terms than I had heard. I didn't make a big deal of it, but I indicated that, inasmuch as this seems to be the biggest thing in the budget, maybe I'd better get to be an expert on it. We had long briefings on Shuttle from stem to stern, and the problems, and so on. Then I talked to Frank Press, and somewhere in this listing, there is no doubt a meeting with the President.
DeVorkin:There are a number of them. Do you want me to look for it?
Frosch:Well, it was probably the 14th.
Frosch:July 14th. See, there was a —
DeVorkin:July 14th, 1977.
Frosch:Yes, there was a June 9th Presidential Review of the '79 budget. And my recollection is that that's just what that was. That is assumed the Shuttle. But maybe not: I don't remember. Let me tell you — I'm not trying to tie it down to one date or the other.
Frosch:I looked at all of this, and I guess my conclusion was the following: it's a little bit like the center conclusion. I am not sure that if I had been the administrator at the time, I would have chosen the Shuttle as it now looks as the thing to do.
DeVorkin:This was in Fletcher's era?
Frosch:Yes, I'm just not sure what I would have done if I had been Administrator then, but since I wasn't. I can't go back to that decision. I do conclude that having gone as far down the line with the Shuttle as we have, and looking at what I understand to be the available alternatives, and looking at what was then JOP (which I see was being discussed) and is now called Galileo, looking at the space telescope, looking at the plans for using Shuttle, looking at the Air Force plans, and so on, it certainly doesn't make any sense to me to do anything but go on and complete it. From a technical development point of view, I can't see any reason why it is not sound. It has advantages. We can argue about other configurations, but it makes no sense to me to go back and start all over again with another round of: "Gee, how are we going to launch things?" I told Frank Press that. I didn't put it that way to the President. I guess it must have been discussed in the budget hearing, this Presidential Review of the budget. The procedure for reviewing the budget, at least in the Carter period, was: you worked as well as you could to make as many compromises as possible working your way up through OMB. First the NASA staff and the comptroller worked with the OMB staff, and then I had meetings, and Al Lovelace has meetings with the senior OMB staff. We, at Al's suggestion, had developed the habit of having people in for lunch. We'd have the OMB people in for lunch, and just talk about things, get them acquainted. You know, this is not a decision session. We're not lobbying you. We just want to tell you about NASA and you can ask guestions; and the real negotiations happen later.
DeVorkin:I saw that in your daily calendar. There's a lot of that.
Frosch:There's an important one I have to tell you about sometime connected with LANDSAT. Then we played a little bit of fast ball in the sense that Al and I found that we easily dropped into the roles that we consciously referred to as "good policeman — bad policeman." This is the classic movie scene in which you put the squeeze on the prisoner to confess by one of the police...
DeVorkin:Oh, yes, yes, yes —
Really talking tough and the other policeman is sympathetic: "You know, I understand you," and so on. I was always the good policeman and Al was always the bad policeman. We fitted those roles, and so we played Saturday morning squeeze ball with some of the OMB people where my line was, "Gee, I couldn't come to this meeting, besides, you know, I was going to talk to the director of OMB," and I'd sit back in the office and read the newspaper. And Al would go over to OMB and say, "well, gee, you're putting me in a very difficult position. I don't think what you're saying is too implausible." (This is the sort of the reverse of our usual ploy) "I'm not sure Bob Frosch will go along with this." We did a little bit of this. They knew we were doing it, but still, it was again part of this business of the administrator position as a final authority: he only talks to kings.
We're negotiating there, and everybody understands it, its part of the way the game is played. We would work our way up through OMB. The idea was at each of these stages, to somehow come to a conclusion that reduces the number of issues. I mean, it is literally negotiation in the sense that we say, "We must have this." OMB would say, "But we've got to have some money some place." and we'd look around and say, "Well, we could give up some of this." And they'd say, "Why don't you give up all of it?" And we'd say, "We can't do that because it's attached to another thing." (You have to have good system sense for this) "It's attached to something else, but maybe, if we do the following: why don't you give us for this package of things, instead of the twenty-five million dollars in the budget, give us fifteen million dollars for the whole package and let us sort it out." We'd go through all of this, and eventually come down to a couple of issues concerning things which we would then sit down and have a shoot out with the director of OMB. The object always being (on OMB's part) to get it all done without taking anything to the President. My object being: "Yes, I'm willing to do that, if we can really get more or less what we want." On the other hand, it's not a bad thing if I actually have to take something to the President. It's good to remind him that I'm there, and it's good for OMB to remember that I have access to him. I'll cooperate very hard if I really feel the President is in a squeezy position, you know, and I don't want to bother him.
He's got other things on his mind, but if not, I'd just as soon go to the President. In this case, the issue of Shuttle was known to be an issue the President was going to discuss. Now, the form in which the issue came to the President was not Shuttle or no Shuttle, because I had already passed the message, in essence, through Frank Press, that said, "That's not the issue. It doesn't make any sense to have no Shuttle. Aside from anything else that's going to cost you more to abolish, payoff the contractors, start again with an unknown, have nothing to launch the Space Telescope etc." Cutting Shuttle out completely made no sense on any level.
DeVorkin:Were you using some of the missions as rationale for that? You almost mentioned the telescope at that time.
Frosch:Yes, yes; in a sense I mean we're committed to the telescope. Everybody wants the telescope, except Proxmire and Boland, but we fought that battle and won. And "you guys want the telescope, don't you?" "Yes," etc. "What will you launch it with?" So there was a lot of that kind of conversation. By then the shuttle was so essential to the national space program (NASA and DOD) that it was worse to contemplate not having it than it was to fund finishing it. The form in which the question came was: we had four orbiters in the budget, and OMB said, "You only need three. Furthermore, you only need one launch site. Why do you need two launch sites?" So that discussion was: three orbiters, four orbiters or five orbiters, and one coast or two-coast launching? Can we save money on Vandenberg Air Force Base?
DeVorkin:Isn't a launch site being built there?
Frosch:Oh yes, but the idea was, we wouldn't bother with the West Coast. Why do you need two launch sites?
Frosch:The President understood this immediately, and the OMB guys — none of them ever did understand it, I don't think. Cutter may have. That's Bowman Cutter, not Elliot Cutler.
DeVorkin:Okay, yes, Bowman Cutter.
Frosch:The President said, "Explain to me why you need two launch sites. Why can't we just work one launch site harder, or build another pad and it will be cheaper, and so on." This is the east launch, south launch thing, eguatorial orbits and polar orbits, and you don't want to launch over land, and so on. He (the President) got it immediately, and we talked a little bit about the angles. That was the first flavor I got of the Naval officer training, because he right away said, "Okay, and what is the maximum azimuth to which you can launch, and why, and what latitudes does this give?" It was clear he had enough recollection of navigation and the orbital elements so that he could ask that kind of guestion. He took the point right away, said, "Okay, I understand. We don't argue about that."
DeVorkin:Is it possible that the OMB people or other people didn't appreciate the fact that you couldn't launch in one direction, then make a right angle turn?
Frosch:That's partly right. Partly, this may simply have been their way of keeping reraising the question. Also, not having the technical training, they may just have been suspicious that they were being lied to.
Frosch:You know, that we could perfectly well do it and were giving them a song and dance because we want two sites, etc. But the President had enough technical training so that he could see the problem right away. I said "You can't make a turn because making a dogleg costs as much energy as launching in the other direction." He understood that, but the OMB guys, I think, had not enough physics to understand it.
DeVorkin:I see. That makes a problem.
Frosch:Which makes a problem. Because they won't believe you. They don't know what you're saying.
DeVorkin:Okay, well then ...
Then we were launched into the four orbiters, three orbiters, five orbiters question. And we were pushing for five, for a lot of reasons. Quantity buys are cheaper. It looked to us as if that made more sense, as we tried to think about two sites, and the Air Force claims for military purposes and what we could see as an economic launch rate and the way the satellite business, communications satellite business might shape up. We couldn't quite see how we could play the shell and pea game with four and satisfy the apparent requirements. Couldn't play it at all with three. We could almost make it work with four. Five looked as if you could really make it work, and still occasionally be able to put an orbiter down for maintenance. We had one other thing in mind that, again, people who have not been in this kind of business wouldn't think of about: an accident. It is going to happen: and it's going to be a problem for the program. Someday, somebody's going to prang an orbiter. They're going to land hard, or something's going to go wrong. or something's going to be odd, in the data. And for a period they're not going to be sure about that orbiter.
There's going to be an orbiter out of action for a period of six months or a year. You've gone to all this money and expense to build the bloody thing: why do you want to save the price of an orbiter (about 7% of the expenditure to first launch)? I think in life of the program terms, but then I don't have to get elected every four years: whereas, the President and his people are thinking of, "What money are we going to spend in our next budget." I'm saying to myself, "By the time we get this system on line, we're going to have spent to a billion dollars. You want the whole operating system to collapse, because seven percent wasn't put in to build enough system? That's crazy." They're saying, "But it costs you an extra $600,000,000 this vear." There's nothing unreasonable about both points of yiew, from the place they're sitting, but that's really what the argument's about. It's a question of: "what's your averaging time." The administrator of NASA tends to have a decade or two averaging time.
DeVorkin:Sure. What was the Air Force's imput at this point? Who were they talking to, and what support were you getting from them?
Frosch:Okay. The support I was getting at that time was from Harold Brown. In fact, it was either this meeting or maybe a somewhat later meeting (I think it was this meeting), that Harold attended, and the President asked him for his view, and he stated his view: he wasn't sure that he could contribute to the question of four orbiters or five orbiters, but he was pretty sure it wasn't three orbiters, and the system should be built. That was an important statement, an important nailing down. I guess Hans Mark was there, too, in the meeting I'm remembering which I think is this one, but I'm not sure.
DeVorkin:That July 14th.
Frosch:Yes, now you can check when Hans got confirmed as Secretary of the Air Force. It may have been later than that, in which case it wasn't this meeting, it was a later meeting.
DeVorkin:Okay, that would help.
Frosch:Now, what else do I want to talk about?
DeVorkin:I'm interested in the fact that during your administration the military visibility or support for Shuttle did become stronger, in the impression of the press.
Frosch:Yes, it was a constant tussle to keep the Air Force hung into the program.
DeVorkin:To keep them involved?
Frosch:To keep them involved.
DeVorkin:They wanted to remove themselves?
Frosch:Yes. They had never wanted to get involved; and I can understand part of it as a straight military psychology that says, "Assets that we take care of ourselves we understand the availability and disposition of. Assets somebody else develops and owns, we're not sure of." Plus there's a long lingering view (which I guess has gotten inherited by now) in the Air Force: "we should have been given the national space mission. We got Dynasaur cut out from under us, and we got MOL cut out from under US. Those civilians who were doing those oddball science things, which aren't even defense, are running the national space program" — there is some of that, I think. So there was always a tussle. The way I see it from outside is that some of the senior people, either because they understood and believe, or because they decided it was sensible policy for the Air Force, really were pushing for keeping in the program. They thought it was important: they could see why it was important, but there was always a group of, I think, more junior officers, although occasionally a general officer, who just wanted to get rid of the damned thing: "If we didn't have to build Vandenburg, then look at all the other money we would have for these things we want to do.
DeVorkin:It is the Air Force building Vandenberg?
Frosch:It's in the Air Force budget, and therefore it was always a great temptation to get out of the program, kill Vandenberg, and somehow keep their hands on that money. Illusory, because they wouldn't: and in fact, I think there was a lot of delay put into the Vandenberg construction.
DeVorkin:Was there ...
Frosch:A feeling that it would all somehow go away. Now, there's one other theme you ought to understand. That is, there's a Titan lobby in the Air Force. That is, there are all the people who build and operate Titans, and they don't see why this magnificent thing with 100% reliability, or whatever the number is, that always delivers for them, and which they control, should be knocked out. "Now, we know we're going to have a Titan; we buy it; we control it; and what's going to happen with those civilians? We're going to want it, and it's not going to be there; and they won't run it right." I'm being sarcastic in a way, but I understand the psychology. From their point of view there's something to it. "We run Titan. We buy Titan. We know where Titan is. We're going to have to borrow this Shuttle from those guys across the river. How's that going to work?"
DeVorkin:Is this the same sort of thing as when you were Assistant Secretary for Naval R and D, and you were faced with the decision as to go to communication satellites from your high-freguency ground-based stuff.
Frosch:It's similar to that. I have to make sure that I can do my job and that I am not totally at the mercy of some other guy. That's a perfectly reasonable point of view.
Frosch:Now, one of the things then that happened in the course of my time there was that we did conclude a formal, let me call it, policy treaty, which I signed with the Secretary of the Air Force, Hans Mark, which said how we were going to handle military demands on the Shuttle. What we ended up agreeing was the following: There is what we will call routine or programmable demands. That is, the Air Force knows, now, that in 1984 in April they are going to want to launch such and such. That's not a problem. We agree, they put a call in and that simply gets scheduled and in the right way.
Frosch:There are what we call national security priority emergency demands — that is, some crucial satellite dies today. Therefore, within 30 days they want to launch another one and this requirement screws up the whole planned program for the Shuttle. After much negotiation, I insisted that the way that is done is: the Secretary of the Air Force, acting as the agent of the Secretary of Defense, calls or writes (calls presumably, it is an emergency) the administrator of NASA and says, "For good and sufficient reason (on a green phone if necessary), I must have this and so." That's the way the agreement goes.
Frosch:Now, there were several questions like: Why do you want to escalate it? There were always people in the Air Force who would say, "But the guy in charge of this business obviously has to call up the Shuttle. It's a military requirement. He's the guy in charge. He calls it up." I told them that out of my Navy experience, I did not want it to be the decision of a lieutenant colonel or a colonel, or even a brigidier general or major general, who says, "Hey, I need it, it's very important; I want it." Because I've seen lots of cases where he thinks it's an emergency, but it isn't clear that everybody would think it was an emergency. If it is that big fat an emergency as to move a whole national system around, there is no reason why they can't get the message to the Secretary of the Air Force, and he can call the Administrator, and they'll agree, if it's that big an emergency. My argument was that if the colonel who thinks it's important on his watch cannot succeed in bucking that message through to the Secretary of the Air Force, then it ain't that scale of emergency.
Frosch:That's all. And people would say, "Oh, but it takes a long time. Suppose the secretary is out of town." You say, "that's bull shit. If it is an emergency, you know you can get him. Don't give me that stuff." "And, suppose the administrator isn't there?" "There is always an administrator available. There is either an administrator or a deputy, or somebody you can call." Then the other issue was, "suppose the secretary of the Air Force says this is important," and the administrator says, "No." What happens then?
DeVorkin:Can you say, "no?" Was that in the agreement?
Frosch:Well, it was implicit in the agreement that he has to ask for it and therefore, the administrator has to agree. I said, "Well, here are two guys who, if they can't agree, either of them can go to the President. And neither of them wants to do that." Everybody agreed, finally, that that was reasonable.
DeVorkin:So you would come to an agreement.
Frosch:We'd come to an agreement, or if it really was a case of the administrator saying, "We can't do this, and I don't believe your emergency matters enough to cancel the only launch opportunity in 75 years for Halley's Comet (to take an extreme one), or the only time in 14 years to do Voyager or Galileo," or something, that was the kind of thing on which there might be a disagreement and perhaps an appeal to the President. That also bears on the question of how many orbiters, how many launch sites, and so on.
DeVorkin:Yes. You didn't want to bring up the possibility that they could always launch on a Titan?
Frosch:Well, that came up later. In fact, my final conclusion was that the system, having forced us into an all Shuttle and no expendable launch vehicle system, because that was the way the Shuttle was devised, had made a very poor decision. The right thing to do, if the system goes successfully and everybody relaxes, will be to rethink the question of what do you launch on what. I'm positive the answer will be that you do launch some things on Titans or on something other than Shuttle. Shuttle is a well-designed system for some purposes, but I'm not sure you want to expend a whole Shuttle turn-around just because you have some satellite that you want to put up that's inconvenient for the Shuttle schedule.
DeVorkin:Right. But how much — I know that the majority of military satellites are in what we call low orbit, a lot of them. At least, the ones I know are. But there must some that require higher orbits.
Frosch:Well, the communication satellites
DeVorkin:Yes, and at this point that required, in developing Shuttle, the development of the launch vehicle from Shuttle that could carry them up.
Frosch:Yes, which the Air Force got the task of doing, what was originally going to be the interim one, because there was going to be a later one. That all had a very complex history, most of which happened before I came on board, and which I never really learned properly as history. People would refer to things. I would say, "What is interim about the interim upper stage?" They'd say, "Oh well, you remember, there was a thing called the space tug," which in fact I didn't remember at all, because I was elsewhere at the time. There was all that history and somehow or other, I didn't know it existed, and so I never studied it systematically, but I absorbed some of it.
DeVorkin:Were there any major design changes for the upper stages or for the Shuttle bay that occurred during your tenure?
Frosch:Not for the bay, but for the upper stages.
DeVorkin:Hmm, so the bay was set.
Frosch:The bay was set, the design was set; the orbiter was being built. Enterprise had been built. It did its approach and landing test the summer of '77.
DeVorkin:Yes, that's right.
Frosch:You know, I went and attended that, and, just parenthetically, was impressed at that by the fact that NASA delegation of authority really worked. I was a guest at that party. Nobody expected the Administrator to do anything. That had been fully delegated. I could stir around and fuss, but nobody really expected me to do anything. I was a very high-level tourist at that test.
DeVorkin:Yes. Let me turn this over.
DeVorkin:Then the design changes were with the upper stage vehicles.
Frosch:The problem was the interim upper stage, which in the fullness of time was renamed inertial upper stage, because they wanted to keep the initials, and everybody realized there wasn't going to be a space tug. so it was going to be in use for awhile. In fact, I don't know its status now. As I left NASA, there was still a question as to whether the development was going to be successful.
DeVorkin:These were necessary for many of the scientific missions, too.
Frosch:Oh yes, necessary for all the geostationary stuff. See, there were three versions, one of which was purely a NASA planetary version. It piled up more stages, or something. I don't remember: I'm getting hazy on this.
DeVorkin:This is something that would fit in the bay?
Frosch:Would fit in the bay with whatever we would design, but that, plus a Galileo, say, would take the whole bay and all the weight. There was a perpetual drum roll of contractual trouble and changes in cost and overruns and technical trouble with that development. I don't know if it's out of the woods or not. That's really how the Centaur guestion came back into the business.
DeVorkin:Okay, we've talked guite a bit about the first contact with Shuttle and its problems. When did you have a feeling that Shuttle was safe and off and running? When did the White House indicate it was satisfied?
Frosch:Well, when the '78 budget, which was decided, I guess, sometime in the fall of '77, was nailed down, then it was clear we were okay. TThen there was the question of the option for an option.
Frosch:I guess it was in '77. It must have been that, out of this discussion of me pressing for five — it may have been '78; I don't really remember. With me pressing for five, and OMB continually saying. "You only need three," we agreed on four. We made the strong point that since there is a possibility that we might want five, if you look at the timing, if we don't somehow run the contract properly and spend a little money (on the Shuttle scale nominal money, ten or fifteen million dollars, I don't remember which) now, then you will automatically have a one year gap, just from the way the budget construction cycles go, and that's going to be very expensive later. You don't want to throw away that decision. "Mr. President, I understand that you don't want to come down on five, but I don't think you want to throw away the possibility of a sensible production run. So, can we write this so there's an option for a fifth?" Well, then we got into a terrible theological argument because OMB was arguing even that an option was more than was needed. Couldn't we keep the possibility without actually saying there was an option? So, I must have invented the terminology. Okay, we'll have an option for an option. We'll keep the possibility, but it won't be formally what is called an option. Now, what the difference was, I never figured out. I think somebody may have invented some legal contractual difference. At any rate, we kept the possibility of a fifth in the thing.
DeVorkin:Hmm. That's really funny.
Frosch:Again, it's the kind of thing: if you're going to build the system, for god's sake, build it right. Don't cripple it because you have some questions about whether it should be built at all. Don't get too damn marginal. The systems analysts always want to optimize things to what they think the situation is, and they don't know the situation well enough to optimize anything. "You really know that you need four instead of five? How do you know?"
DeVorkin:You indicated that plans are afoot to modify the basic philosophy pro-Shuttle in that it would replace expendable launch vehicles with a policy that allows for ...
Frosch:This was something that came up again. In 1980, I guess, we were talking about that. It was something Hans Mark raised, and frankly, before that, nobody even wanted to raise a discussion of the subject, because suddenly there would be too many people running around saying, "Aha, they're sure the Shuttle won't work!" People draw conclusions that are convenient for them. If they don't like the Shuttle, everything you do, no matter how sensible, is evidence that you know something you're not telling. What are you not telling us? What did you discover, and so on. The press is particularly good at that, because, poor things, they never understand what's really going on, even when you explain it to them. They are deeply suspicious. Not so much the trade press, but the general press. Then you have Congressional problems. If we, after going back again, asking for more and more money for the Shuttle, if we suddenly said, "By the way, we want some money to build some expendable launch vehicles," we would have been in deep, if irrelevant, trouble. So you frequently do not do what rational planning would tell you to do, because there are political obstacles. You just know what the reaction is going to be. You've talked to people. You know the reaction. You can't go in and say, "Look, the sensible thing is to do thus and so," because other conclusions will be drawn.
DeVorkin:So you found yourself committed to the Shuttle, and thereby, committed to manned space flight.
Frosch:Yes, yes, but that was all part of the package.
DeVorkin:So maybe, it's at that point where we could talk about manned space flight versus unmanned.
Frosch:Yes, let's take a break first.
DeVorkin:Okay, go ahead.
Frosch:The question of manned space flight — I guess I started the consideration being neutral. That is, not accepting the Shuttle just as a manned thing; accepting the fact that clearly there were, in the long run, things people would want to do in space; and noting the fact that there was a very strong remote control capability that hadn't existed in an earlier stage in the business, and a strong extensor capability. That is, a capability for human beings to be part of the control over something that was a long way away. I am here, but my eyes, ears, sensors and hands are there, the direct hands-on capability. This was an interesting set of questions. Incidentally, there was a question, a presidential question that came up, I think, in this first discussion: why can't the first launch be unmanned?
DeVorkin:That's very interesting
Frosch:To which my answer was: it could be, but it wasn't planned that way; and it would now cost an awful lot of money to make it automatic. That mode wouldn't help much in learning how to fly it. It had to be flown, so I don't think it's worth revisiting that decision.
DeVorkin:Was he saying that out of concern for the astronauts' safety?
Frosch:Yes, I think it was that somebody had reminded him that we had always flown the first flight of a vehicle unmanned before. He was being sensitive to the political problem of possibly killing somebody on the first flight. I think safety, yes, but behind the safety, the whole question of what the programmatic and political difficulties were of even exposing astronauts the first time.
DeVorkin:That became a red herring just before the launch.
Frosch:Oh well, the press kept bringing it up. One of the points is, people forget conveniently, and they forget the fact that it is true we had always had unmanned launches before the manned launches, but as George Low reminded me, there was at least one case where the unmanned Apollo blew up on the pad, and they went ahead with the manned one two weeks later, anyway. I think it was a Gemini case, but I don't remember.
DeVorkin:Hmm, there was also the Apollo fire.
Frosch:Yes, and so on. There was all that history that people have forgotten. There's a tendency on the part of the press, in particular some of the press, who aren't so heavily involved to regard Apollo as having always been always smooth with no problems or difficulties. Why can't you do that any more? Just beautiful. That kind of thing.
DeVorkin:Do you think we'd be doing a justice to the whole program by presenting a more objective view of the Apollo fire here? We've been wondering whether we should not display the charred capsule and discuss all the problems that had gone along with it.
Frosch:It would throw NASA into a state of shock if you do.
DeVorkin:Do you think it would?
Frosch:Yes. NASA has convinced themselves that that would be very destructive for the program. I don't know whether it would be or not.
DeVorkin:Whom should we talk to at NASA to come to some understanding of that?
Frosch:You might want to start with the Administrator.
DeVorkin:Okay (laughs). We don't want to create any problems. We just want to say, in a way, that's more objective history.
Frosch:I think you will find that the manned space flight portion of NASA is dead set against it.
DeVorkin:Okay, well, let's continue on.
Okay, so I started with that set of questions, and have come away from NASA, still with a lot of ambivalence about NASA's perception, or the manned space flight part of NASA's perception of why space flight should be manned. See, it's clear to me that, if we really do want to explore Mars in the end, we're not going to do that with unmanned or remote vehicles. I mean, you can do some things, but a real exploration is going to require men on Mars to control what they're doing, simply because the time delay begins to be long enough, so you can't do it sensibly unmanned or remotely controlled, I believe. On the other hand, I came away totally unsatisfied with the ideas that the manned space flight community has about what it is people should do in space. My problem with the Jerry O'Neill kind of thing remains: why are the people there? You know, why do you want to colonize empty space? And I don't believe any of his reasons. They keep talking about Columbus.
Well, you know why Columbus went where he went. He thought he was going to the Indies and getting gold. Very straightforward. You know why the Pilgrims came. They were running from something; and you know where they were going. They were going to a place they believed they could farm and live. They weren't starting from scratch, and then after them, even though people were running, they were running to good land, the Oklahoma land rush, all the rest of it. They were reproducing in a new place what they were doing elsewhere. I'm not sure all that applies in the space case. We start out with an exploratory reason and we go on to the question of energy and material. I've recently taken to asking people: "where do you live?" And I say, "No, no, no, I don't mean that: do you live on the Earth or in the solar system?" But I mean it in the sense of habitat and useful reasons for using resources and so on, "I want to go out there and explore."
DeVorkin:Because it's there.
Frosch:Because it's there. A perfectly sensible reason, and besides, it has generally paid off. That has its sales problems. I'm not even sure I want to try to convince the American people to pay for that. People have traditionally not paid for something an explorer has done, but that's a disgression from what I want to say. I don't know whether I've said this before. Let me go through it again, even if I have. I ended up asking the following kinds of questions a year or so ago. I said, "You guys mean to tell me that what you propose to do is select again and again out of eight or ten thousand applicants twenty-five or thirty of the absolutely best people you can find, with doctor's degrees or equivalent in all kinds of fields, postgraduate experience, trained, honed spectacular. You're going to put them through two to four vears of post-graduate astronaut training. You're going to send them up there, and you're going to say, "Okay, put on a space suit and bolt bolts on a structure eight hours a day."
Frosch:That makes absolutely no sense to me. You're not really going to take people out of a hiring hall and put them in a suit to do that. It's too expensive. You're not going to put structures together with people in suits. You're not going to design most of what you do to be done with automatic machinery. You know, for purposes of work in space, you really design things to be done by machines, and the real reason you have pilots and astronauts in the Shuttle is because you're going to want to elaborate things that are beyond automatic machinery. You're going to want people who are doing honest-to-god experiments which for some reason are hard to control from the ground. The pilots in the Shuttle are, as much as anything, being sent up to keep the Shuttle safe. You're not building the Shuttle to keep them safe. You have too big an investment in that system, and you know about pilots. Pilots can rescue systems in situations that you wouldn't have dreamed of putting a program into the machine for.
Frosch:Now, that's the whole story, especially in the beginning of the system. That's the real reason for sending the pilots on the first flight. You're not sure you know how to program the flight, but you know John Young and Bob Crippen, given an unexpected situation, would be better than your machine program, although not to do the routine controlling. The machine can keep the thing at the preprogrammed alpha better than John can, but what happens when it does something screwy that isn't in the flight program? He's got a better chance than a canned computer. The same is true of elaborate experiments, and so on. We proved you can run the space telescope from the ground. You really don't have to have astronauts in that. It may be that when we have elaborate antenna forms it will be very sensible to have a couple of people there to do some maintenance things, because the robotics may be too complicated. In the end, it may be simpler to have a robot up there that's controlled from and instructed from the ground, and cheaper.
DeVorkin:How do we know that the space telescope can be controlled completely from the ground? What experience do we have? Is it IUE?
I'm really thinking about IUE. We have an experimental proof that you can run a telescope from the ground — IUE. Now it may turn out you can't quite run space telescope, but it seems to me you can — that we really do have a piece of operating experience. We have enough satellite and spacecraft operating experience. So for that, we think we can do it. There is no time lag problem. There is no difference between running the space telescope in orbit from the ground, and running the telescope in the dome over your head from the room down here. All it is is the communication links, and the only question you have to worry about is, have you done the communications links right? That seems all right to me. You want people for the unexpected. If they are inventing on the spot, if they are deciding what to do on the spot, and they can't decide it from somewhere else. These decisions are going to get harder and harder as the robotics gets better and better.
I don't want to use the term artificial intelligence. I iust don't want to get hung up on the philosophical business: is the machine intelligent? I don't care, but as the amount of human interference with what you've told the machine to do gets smaller and smaller, then the question of what are the people there for, in terms of work to be done, is going to get more and more difficult. My conclusion is that, just from the point of view of doing work, quite aside from exploring or colonizing, that within easy reach of the Earth (by which I mean moon time delays, not Mars time delays) then you're always going to have a difficult problem deciding between machines and people. Unless you go to the back side of the moon, where you have a different kind of question. Where should the people be? Should they be here, or in a satellite? And you're going to be making the decision, not on the basis of: "people should be there," but on the basis of what's the best way to run the system. This is a very unpalatable view to the heirs of Werner von Braun, you know, because they have a vision of doing it in a particular way, a vision which couldn't be contradicted very well 20 years ago, because you had no alternative.
DeVorkin:How many of those people are still in NASA? How strong is NASA still in that vision?
Frosch:Quite strong. Quite strong. Chris Kraft is a very strong example. Lots of people are there. Now, they have some counter arguments about what people can do, and they don't quite believe in the robotics and so on. They are pushing people out into space. I think people are going to be pushed out into space by all sorts of reasons. And I'll come to my peroration in a little bit. But I think it is counterproductive to invent artificial reasons why people have to go out in space, because they are transparent, and people think you have no reasons if you produce artificial reasons.
DeVorkin:But you say in the same sense that the pure adventure of exploration or knowledge for its own sake, however.
Frosch:I think exploration is sensible, and I understand it. People want to go walk around on Mars. It isn't quite just because it's there. It's because they realize that, at least in the current technology, and the immediately foreseeable technology of: I am here and I'm looking there, it's not going to give them the same realistic feel as standing there, even in a suit. This is complex. There was a recent big thing in which Sylyia Earle, the oceanographer, went down in a hard suit to a very deep depth. Now, I looked at that kind of hard suit. And I don't see why she made it in the shape of a suit, rather than either sitting in a minisub with arms, or sitting on the surface with arms on the end of a cable.
DeVorkin:With a TV.
Frosch:With a TV. See, this is getting more difficult than it used to be. When Al Yine and I invented, pushed the idea of Alvin —
DeVorkin:Yes. Yes, right, I remember.
Frosch:Right. Then I made a computation, and said, I don't know how to transmit the band width of a man, so I've got to put him there where he can look out through the plexiglass. Now I can come closer to transmitting the bandwidth of a man.
DeVorkin:In other words, all of his sensors.
Frosch:Not all of his sensors, but all of the senses he can get through a plexiglass window.
Frosch:On the other hand, it is not clear that moving a remote sensing sub or a remote-control sub around quite yet would have found the Galapagos rift hot vents the way Bob Ballard and Alvin did, but the technologies are getting closer and closer. The way I put the problem is: it's not the question of whether you use people. The question is, where do you put the people? Let me give you a strong terrestrial example. I don't see why there are all those coal miners in mines. I think we now have the technology so the coal miners could sit in an air-conditioned building on top and run variations of the same mining machines down in the mine. This would have two advantages, which pay for the technology. One is, you don't have to worry about the safety factors of miners most of the time. You do for some, because somebody's going to have to go in and repair machines, but that's a special case. And the other is, you don't have any black lung disease. Just the insurance premiums on black lung disease and miners, injuries really ought to pay for the system. Now, the counter example is, we still have human divers in the North Sea and in the Texas area. That I believe will gradually shift over as the controlled robotics technology improves. Right now, it's an economic problem of slavery. I don't mean that in the literal sense, but I mean you can hire divers to put their lives at risk, in some cases cheaper than you can make machines to do it. In fact, you've got that problem with astronauts. If it's just a matter of paying people to go be astronauts as opposed to buying machines; you've got volunteers. It's easy. Lots of people will go for three squares a day.
DeVorkin:But aren't the support system —
Frosch:But the support system makes it ferociously expensive. And when that crossover begins to come between what it costs to support people, and what it costs to provide, to some extent, people-equivalent robots, I don't know what happens. Then, of course there is the problem of the unexpected. You've got robots, and you've got robots that you can control, that repair robots, but somehow, you have the feeling that the system may not converge quite right, and you want somebody there who can say, "Oh, is that the problem?" and fix it. Now, to some extent, I can do that with the little machine with the hands and the TV, and I look and I say, "Oh. " and then send it over and it twiddles.
DeVorkin:Because you're talking about more and more sophisticated machinery in space: therefore, the person up there has to be more a specialist to understand them.
Frosch:Yes, well, now, what is it the astronauts are going to be doing? Well, right now, you've got astronauts who are going to be controlling the arm and take satellites and so on, and going out and twiddling with things, and fixing the spacecraft, and so on. But, right now the technology exists so that you can computerize the control of the arm, and instead of doing all the fancy things, you can, in effect, tell the computer, tell the arm: grab that, take it out of the bay, and move to XY rho theta position, and tell me when you're through. We can do that. There are systems like that. That kind of work you can remove, but it's the case when an astronaut looks in the bay, and there's something wrong in the bay. If the arm followed its instructions it would get tangled up in a cable that isn't supposed to be there, but is supposed to be dressed down along the side, so the astronaut must take care of that kind of unexpected problem. There is the solar experiment where it pays to have somebody to sit there and look at the sunspots the way he wants to with an optical telescope and not a TV transmission and twiddle with the experiment. These cases are continually going to happen. To some extent, you want some people because they will see things you didn't program yourself to see through the system.
DeVorkin:That I think is ...
Frosch:You see, that's very important, the unexpected. That's what's important about having people go to Mars or the moon or whatever. Unfortunately, Apollo didn't demonstrate that very well, because it was so hard to get them there and so dangerous, and everybody was so scared, that they really didn't have any time to go wandering around looking for funny things, which is what they really should be doing. You really would have liked to have been in a position to put them in the jaunting car and say, "Okay, fellows, go have a look around and see if there is anything funny around." On the moon, the answer might well have been, "Hey, there's nothing funny around here. On Mars, the answer probably is certainly going to be something very peculiar: I don't understand this, and I'd better turn over this rock.
Frosch:Do it and not ask Houston, "Is it all right if I turn over the rock." and Houston has to go into consultation huddles while they decide whether the astronauts turn over the rock (laughs). As long as you do that, then you don't need the astronauts up there. You could have them put them in Houston.
Frosch:And send a machine with arms.
DeVorkin:Well, that's what they did with Viking.
Frosch:Right, and there is no understanding at Houston that, as long as Houston tries to run the astronauts, there is no sense in having them. It doesn't exist yet.
Frosch:If Houston is technologically magnificent, philosophically, I find them a little backward. Meaning by philosophy, this kind of issue: if you control them from the ground that tightly, you don't need them there, almost, or close to that. They are really telling you what their problem is, and you tell them what to do about it. The rule usually is, the guy on the spot knows better than the guy who's not there. If the structure were that the astronauts were saying to Houston. "So and so and so and so: I need a computation like the following: Please tell me what the computer says," it might make sense. But even that isn't necessary any more. There is no reason why the guy on the moon doesn't have his own control console for the machine in Houston. (laughs) You've got line of sight video, what the hell. Let them run the machine. Why does he have to talk to some guy in Houston who sends somebody in the backroom to run the machine. That's a technological change from the time of Apollo. You didn't have microelectronics, really, in Apollo. You couldn't give the guys on the moon a micro with its own buffer to run the machine in Houston. That technology hadn't been invented. While the guys at Houston are using that technology in their own simulation and control, they've got a mental block about letting the astronauts, even thinking about letting the astronauts on the moon, do it their way. NASA is so clutched up about having a death in the family, and everybody saying space is too dangerous. Fifty-thousand people a year die on the roads and we're putting two people in danger. There is a difference, of course, because fifty-thousand people die on the roads in private in a sense, not on national TV, but somehow, the system has to get over this difficulty.
DeVorkin:Did you ever have sessions with people at Houston on this kind of topic?
Frosch:Well, not so much. We had some sessions with people from Houston in which we discussed it in headquarters. And I think I made them very uncomfortable, but that's about all.
DeVorkin:Who were these people? Do you recall? Chris Kraft?
Frosch:Oh, Chris Kraft, and mostly the Shuttle people would be involved.
DeVorkin:Yardley, or anybody like that?
Frosch:Yes, John, Chris, Bob Thompson. Houston, hearing about these discussions and these views, and hearing them directly, was very worried about them. I wasn't of the true faith.
DeVorkin:This was early on in your administration?
Frosch:Oh, I think throughout. Early on, they didn't know. Later on, they began to worry about me. I wasn't of the true faith. I don't really know. They are always very friendly and Chris has always been a great support: but I think he regarded me as a little bit strange. Still does, I would guess. I don't know what he would say. I never asked him.
DeVorkin:It's a very interesting complex question. I mean it pervades popular culture more than, maybe, popular culture is aware: because listening to your talk, I was reminded of a scene in "2001" which is, of course, quite old. But it's still —
Frosch:Yes. Incidentally, I've never seen it.
DeVorkin:Oh, you've never seen it.
Frosch:I've never seen it. I keep being in the wrong place when it's available.
DeVorkin:Well, there is the scene.
Frosch:Never tell Arthur that I haven't seen it (laughs).
DeVorkin:Okay. I guess I can't ask you what you thought of a scene, but there was a scene where the computer indicated a malfunction, or predicted a malfunction, and before the astronauts could go ahead doing anything, they got permission from back on Earth to do this and that and the other thing. And I always thought that was a bit overloaded, but it represents NASA's philosophy.
Frosch:It does represent Houston's philosophy.
Frosch:Well, I think of it as that, because NASA's perhaps beginning to grow away from that a little bit, but it hasn't yet.
Frosch:Now, let me go on to the extreme thing. This is something that I pushed a little in NASA. I'm still pushing, and you ought to have it. This happened in the spring of 1979, the issue was raised by Richard Muller, who's at Berkeley. He's a physicist, who got an NSF prize — I forget which prize — for some brilliant innovations. In fact, he made three, in three different subjects in the course of two years, the rubber telescope being one.
DeVorkin:The rubber telescope. This is the ... ?
Frosch:Yes, the multiple mirror.
Frosch:The multiple mirror: not the distortion telescope, but the multiple mirror closed loop control telescope, and a couple of others which I don't remember. And about that time he wrote to Frank Press and pointed out that in each of these cases, including one with NASA, the system wouldn't have anything to do with them, and he had to fight his way through getting somebody to listen to him. He suggested that the system generally, not NASA particularly, the government research support system generally, was not very receptive to the offbeat idea that didn't come from the right place. That's a view I'm sympathetic with. I think that is a real problem. I have this theory, in fact, that great inventions are always made by slightly wrong people, for reasons I could articulate. So you have an orthodoxy problem and a peer review problem. One of the things I set out to do was to find some ways in NASA to make the system more flexible. I guess we did a couple of things. One: by main force I made sure there was a pot of money that an Associate Administrator or a Center Director could use arbitrarily for things that looked like a good idea, to be reported on later, with a corrective system that says, "Hey, if after a couple of years you are solving mundane problems, you're not going to get so much more money: and this guy over here who made four different wild bets, two of which came out, is going to get more: we're going to bet on him." That system I think is still operating, not with great guantities of money, but enough so a few interesting things can be done.
DeVorkin:Now, this letter went to Frank Press. Did it come directly to you?
Frosch:He wrote to me. Yes, Frank sent me a copy. Rich Muller wrote to me, and I talked to him. I had him in and talked with him.
Frosch:And so we did that thing.
DeVorkin:What was the name of the program?
Frosch:I'll be dammed if I know. I don't think we had a name.
DeVorkin:But it was a discretionary fund that —
Frosch:It was a discretionary fund program. That center directors had. You had. Okay. Yes. And now, the other thing was: I said, "Gee, we ought to do some things that are a hunt for new ideas." This was a John Naugle task. He started, I guess, early in the spring of, maybe in the fall, of '78, inventing a way. John invented a way to systematically find some new ideas that worked very well at least once.
Frosch:What it was, now how did it go exactly? You started in the fall, and you wrote to a lot of people, and you told them what the structure of the program was, and you invited people to be part of this process, by name. Some from in NASA, some from outside of NASA. When you wrote to them in the fall, you said, hey, we're going to have a two-day meeting in March, or February, whenever it was. Everybody who can come, come, and what we're going to do is look for new ideas. Any NASA subject. By the way, if you have anything that you want to throw on the table, write me a letter and I'll circulate the letters, and maybe we'll try to put them together into a list of what will be before us in March. Some of the people came. There were letters, and some of the letters were very mundane ideas we had heard before, but there were some ideas we hadn't heard before. Out of this, by inviting people, we structured a week at Woods Hole in June. Because we had had the preliminary meeting and the preliminary structuring, we were able to organize the people so that the day they arrived, we could have a morning session, and then everybody was organized into groups by what they wanted to work on.
DeVorkin:This was '79?
Frosch:I think this was the spring of '79, the first time we did it. We did it again in the spring '80. It was not quite as successful. But it really worked, to different degrees. We had a small group of astrophysicists. Frank Press's son, who's a professor of astrophysics at Harvard, was one of them. Rich Muller was another, and a couple of other guys, very good guys, arrived, and they had ideas, and in a week they made a far out program of instruments they think should be built some day for various purposes. Very large-scale interferometry in space, very good stuff. The second year, by the way, we had an interesting set of solar things done, including a thing called centurion, which is to be a very simple, very long-lived satellite. It is supposed to live a century to take certain long time baseline data on the sun. You can't interfere with it and turn the program off because it is inconvenient, and it reminds you that it's there, and you can pull out the data. Somebody started this because he said the British started a program to watch the sun indefinitely in 1880, or something, and the program got knocked on the head ten years later, and we sure wish we had the data. That kind of idea.
DeVorkin:I remember reading about that. That broke through. People talked about that.
Frosch:Riccardo Giacconi, before the first one of these, said, "Gee, I'm kind of interested in robotics. I just don't know what's there, but I want to come to the meeting and talk about robotics."
Frosch:Yes, he kind of stumbled onto this, and he spent a little time worrying about where some of this was going, and what people were capable of, and what they were doing. We found ourselves in a group that included me, because I insisted on the right of coming to the party, and John insisted on it. He thought it was good for Administrators to do this.
Frosch:John was in the group. Carruthers, who was our materials guy.
Frosch:Carruthers, John Carruthers. And let's see who else? Barney Oliver, and, gee, I'm losing names at a mad rate. George, a tall astrophysicist form the Smithsonian Astrophysical Observatory.
Frosch:George Field, of course, was there part of the time. Who else? Sadin, Stan Sadin, maybe.
DeVorkin:I don't know: I don't think so.
Frosch:Anyway, we sat around and said, "You know, there's something in this business of robots, and also Riccardo had read some of the stuff by Von Neumann on Von Neumann replicating machines.
Frosch:And Barney Oliver knew that business, and we started to kick this around in the spirit of: "Is there a pony in this pile of manure somewhere?" We found ourselves talking about the availability of resources and so on. At the end of three days, we had designed the following concept: the availability of materials for space exploration via self-replicating machines on the moon. John Carruthers would say they should be on an asteroid, but it doesn't matter.
DeVorkin:Let me turn this tape.
Frosch:Anyway, this became the Administrator's hobby. But the concept is based on the following kinds of things: Let's ask the following questions: I put a solar cell on the moon and it starts to collect energy. How long does it take a solar cell given real system efficiency, (and we threw in 10 to the minus two just to make sure we had some slop factors) —
Frosch:To collect enough solar energy to make another solar cell, extracted from whatever material is there. You scoop it up, you refine it, you do all that. It turns out the answer is of the order of several days. Not a very long time. And so we made some computations that suggest the following: You build a Von Neumann machine, which is a small factory run by solar energy, and let's say it weighs 10 tons to a hundred tons, say a hundred tons. Put a hundred tons on the moon, and you give it this solar cell conversion efficiency with an extra ten to the minus two, and you say, "It's a machine that sets out to build another machine just like itself," and away we go. And you then say, "What do I have in twenty years?" What I have in twenty years is ten to the seventh machines, its exponential growth. After all. I've used maybe one-hundred kilometers by one-hundred kilometers by two meters deep of the moon, and what do I have? Well, I have a pretty general purpose industrial facility run by solar energy. In fact, if I throw away the industrial facility and connect up all the power systems, I have a — I don't remember the number exactly — a large multimegawatt, a gigawatt size plant.
Frosch:Now, there are lots of interesting problems, like can you in fact converge on the machine which makes the machine. I made a heuristic argument which I won't detail, that says, "Well if I can't converge on the machine that makes a machine, I can sure converge on a system that takes damn few people to control it, including a guy who comes along and says, "Oh, there's a rock on the track. That kind of thing, or controls the robot that looks around for that kind of thing, and so on." But at least, if we don't have a theorem of closure, we have a plausible theorem that you can do it with very few people. Now, why do you want to do that? Well, there are two divergent lines of thought. What we concluded in that brainstorming session, the way Riccardo said it was, "We can now explore the planets free." The idea being that the economics is very straightforward. For whatever it costs you to put one-hundred tons on the moon, plus some support costs, you now have access to all the materials there are in the moon, paid for with free solar energy and a single investment, which means you have an infinitely large, or an extraordinarily large, return for finite investment.
DeVorkin:If you're only willing to wait ten years.
Frosch:If you're only willing to wait. If you're willing to wait. It's exponential, its pseudobiology. What you've done is built a pseudobiological system. Now, there were lots of interesting digressions, like Barney Oliver pointing out that you might have an A-B system in which machines of type-A build machines of type-B, which build machines of type-A, and that might be better: and all sorts of digressions.
DeVorkin:Build a better lathe technology.
Frosch:Yes, we've looked at that question, too. So this idea came out of the study. Incidentally, in the summer of 1980, I guess it was, there was a summer study, of the Ames summer study types, that looked at some of this and drew some more conclusions: and I think there is a publication coming out. Now, the other line of thought other than: gee, the NASA program can now sit on the moon and do what it pleases, led to my question of: do you live on the earth or in the solar system? Because, if you do this, then you have an economic program that says, "The human race has access to the resources and energy of the solar system, not just the resources and energy of the earth," on a basis which makes some kind of economic sense. That is, it's not a basis that says, "First you put in $150 billion, and then, maybe, you will have some kind of funny self-sustaining colony." This says I do all my work on the earth, and then I know what it costs me to put this machine on the moon, so I know my investment, and the whole thing runs from there. (I later learned that we were retracing a line of thought followed by Freeman Dyson some time before. See e.g. his book, Disturbing-the Universe.) Next we asked ourselves an interesting question: if on the moon, why not Arizona? Everything I just said about a self-replicating machine is also true on the ground. The answer seems to be: because they will let me do it on the moon. We don't know how to handle the equivalent of a Midas economy. That is to say: if you can build a machine that will sit there and build machines like itself, then you can reprogram this machine so it will build nearly anything like electric razors or whatever. This neatly splits industrial production from the distribution of wealth. It is the extreme answer to the labor theory of value. It's value with no labor at all.
Frosch:So you're left with a very funny economic question: how do I run a system in which a large proportion of the goods are free: at least, after I have paid my investment back, they are free: and still run a human race. You know, how do I distribute all these goods? What does the economic system look like? How do you deal with the agricultural part of the economic system, which isn't run by this, or what do you do with it? Our feeling was that as we look at the technology over the next twenty or thirty years, forty or fifty years, the technology is certainly going in these directions. Who the hell's going to look at the social and economic questions? I tried this idea out on many people in the months after we cooked it up, put it in speeches and so on. Most people just kind of looked at it and said, "You can't build that machine." I said, "Well, I think we can. Suppose we can." Economists would say, "You can't do that, for no reason other than that's not what economic theory says. I had a few people say: "You can't do that, because —, "and then there would be a long pause, and "I'll talk to you tomorrow." And they usually came back and said, "Well, I just feel you can't do that, but I can't find a reason why." It must violate something (laughs). It doesn't violate the second law of thermodynamics, by the way. You are taking in heat from the sun and putting out heat, and doing work and it's a perfectly reasonable thermodynamic system.
DeVorkin:Did somebody suggest that?
Frosch:No, but there was some initial reaction that it must violate something.
DeVorkin:Even if it did, I wouldn't worry about it.
Frosch:Well anyway, you see how this complicates the human question.
Frosch:Because now I'm saying, you know, I can even run my industrial economics, in principle. I think I can get to the place where I hardly use any people at all. We're going in that direction. And the economic and social implications are potentially severe. There is the other thing. History tells you that every time you have built a labor-saving device like that, what it has done has generate a lot of other things that you wanted to employ people to do.
DeVorkin:Sure, but there's that lag.
Frosch:There is that lag. There is also a guestion about which people. We now seem to have divided the U.S. economy into people who are in it, and people who don't have the talent to be in it. Either the talent or the education, or the something, whatever it is: they don't have the skills. And we don't have any use for unskilled people. You don't hire people to use picks and shovels very much any more.
Frosch:Watch them build things in New York City. In fact, it is rare to see a guy with a pavement breaker. You see them, but mostly, what you now see is the construction worker, or machine operator, standing in a big machine which does the work and does it at such a rate that ten guys with picks are no longer necessary.
DeVorkin:Right. To get back to the guestion of doing science today as opposed to ten years from now, out in space in different ways. There is also the lag time that I see in the scientific community. We already talked about this: you know, in manned space flight versus unmanned. One of the biggest arguments has always been: well, if we can just get the Shuttle going, then down the road there's going to be far more potential for getting into space to do scientific experiments. But the people who want to do the science now are not satisfied.
Frosch:I had some arguments with people there, too. Some of the planetary people.
DeVorkin:Did this arise at these summer studies?
Frosch:This arose in various other discussions, not so much at the summer studies although it occasionally did in some of those discussions. Some in discussions of what we should do about various planetary and space programs. You know, there were a group of people who would say, "Look, a Titan can lift as much weight as a shuttle: and we could put Galileo on a Titan. What do we want a shuttle for? It doesn't do anything for us. We can build a bigger expendable launch vehicle, and I couldn't understand why they were so narrow-viewed, because I could instantly see at least one thing it could do for them right away. It means you can take up several packages, and you can store them in space, and you can assemble them at you leisure, even without robots and fancy machines. That the astronauts can do.
Frosch:What does this do for you? The first thing it does for me is it says, "If we're going to build another Voyager, I can now put all the fuel I want in the Voyager." You want fifty tons of fuel? You've got a weight and balance and moment, and delivery and costs in shuttle flights, and so on. But there's no reason why you can't put fifty tons of fuel, if it's worth it to you. In fact, even with the combination we haven't got of cheap expendables and an occasional Shuttle flight, you take the crate of eggs that is the instrumentation up there very carefully. Maybe you go up to geostationary, I don't know. You carefully hand tailor this thing. Separately, you're just lofting fuel into space. And then you're assembling this great, monstrous kluge out there, which no longer has to fold into a Shuttle bay or into an expendable nose cone. If you want a big machine, then you can have it. Well, even if you want a lot of simple machines, but a lot of fuel, so they can stick around some place for a long time, that's clearly not twenty years off. If we had the money and the shuttle flights, we could begin that experimentation next year. It's large ground assembly and the self-erecting characteristics that are not necessary. If you would like to bring up your fancy instrumentation that you want to have calibrated right there on the machine, you can do that.
DeVorkin:How did the JPL people react to those arguments?
Frosch:JPL in the past few years has been so beleaguered with regard to whether Galileo was going to survive, and so on and so forth, they really didn't have any patience to discuss that kind of thing. They are too worried about whether the planetary business will survive. Now, I had another argument with them about science, and I'm delighted to see from something in the newspaper or in "Science, I guess it was, that this bore some fruit. I think Andy Stofan took it seriously. This was the following: "Look, we're going down a line in planetary exploration which is increasingly difficult to support politically and budgetarily. That is, each mission is more expensive than the last, by a large factor. Now why is this?" To begin with, we're doing harder and harder things. I mean, it's just harder to go to Jupiter and Saturn than it is to go to Mars.
Frosch:And we're doing fancier instrumentation, more and more complex: but this means that it comes in bigger lumps. You guys are buying six hundred million dollar missions. That's a particular way of packaging the problems. Is there some other way of packaging scientific curiosity into missions that have a different kind of characteristic? Should we really be putting the particles and fields stuff on the planetary observation satellites? Is it really right that we are somehow saving some housekeeping by doing that? Should we have a whole sequence of cheap things? Is there such a thing as a sequence of cheap things that only go out there to look at particles and fields?
DeVorkin:Yes, I see what you mean.
Frosch:Those don't have to have all this great precision of navigation and so on. Is there some other way to package, to regard this as a data-taking and an investigation pattern which is packed up in missions in a different way. Now we're only going to get one mission in the next three years: and so it's a Goliath. Is that a self-defeating line? Maybe we could go back to the program packaged little mission business. There was a whole series of scientific satellites which was a program: we spent so much a year on the program, and put up lots of satellites.
DeVorkin:Was it the OSO series, or the ...
Frosch:There is a name for the whole program.
DeVorkin:Not the Explorer series?
Frosch:Yes, the Explorers.
DeVorkin:Oh, the Explorer series. Okay, that was forty or fifty satellites.
Frosch:Yes, is there some way to take an Explorer-like philosophy in planetary. You know, the smaller, discrete experiment in which there is a level of effort, and we get a lot done without going through these paroxysms of giant project development and budget defense, and so on and so forth.
Frosch:Is there a way to smooth the program out? I gather from something I saw in "Science" that this is now again seriously being looked at. I think I started that line. Everybody's immediate reaction was. "No, no, no, you have economies of scale," and so on and so on. Economies of scale may kill you. What is it you are trying to minimize? Is it cost of this year's project, because it's called The Project: or is it the cost of the next ten years of planetary data-taking and exploration? You know, we had started down this line of a thing called The Project, to be the most economical package for looking at Jupiter. Well, maybe we're not looking at Jupiter. Maybe, we're really looking at the solar system, and the things that you want to look at are more general than on Jupiter, for one thing. It was just as vague an idea as I'm saying. I couldn't specify it, but I think we pushed people into looking to see if there are other patterns of exploration.
DeVorkin:Was this kind of a feeling that you have an outgrowth of the problems of maintaining successfully, or unsuccessfully, solar polar and VOIR?
Frosch:Yes. You know, this was a continual effort. You had these big name things that were expensive, and targetable in a budget sense, and every year it was, "gee, why don't we cancel?" Now, partly that was because of the political battles over the telescope and Jupiter orbiter probe, and Eddie Boland's having lost a vote and being annoyed over that.
DeVorkin:We discussed that.
Frosch:Yes, yes, the feeling of political people that they're not quite sure why we want to go look at Jupiter. I mean, it's exciting and it makes lovely pictures, and the scientists are terribly happy, but, you know, that's a billion dollars to go do that. In fact, remember that there's a battle inside the scientific community. There are the NIH guys who say, "They spend six hundred million dollars for that, and I can't get two-million dollars to do my brilliant whatever- it- is." Science comes in different scales, and the people who are doing it in smaller scale don't understand why somebody's doing it at larger scale when they could do so much more science for the money. This is a sort of "bang for the buck" theory in science, which says all forms of science are the same. We should buy the cheapest with which we can make progress. They don't present it that way, but there is such a philosophy.
DeVorkin:I know there is. How is it to work in a philosophy like that? It must be very frustrating, especially when you see within the scientific community these differences of opinion within JPL itself. To what extent were you involved in any problems, internal problems that JPL had in deciding for VOIR or Galileo, or ...
Frosch:Not much: not much. Only as they would be reflected through what Noel or Tim Mutch ...
DeVorkin:They were directly ...
Frosch:Well, I don't know how directly involved they were, but only reflected in what they would know about that. I would occasionally hear something from Bruce Murray in conversation, but mostly not.
Frosch:I think I have a specific time appointment to make.
DeVorkin:Okay, good. First of all, I want to thank you for this session, and we'll go on to another one.
Yes, interesting, this one.