TAPE 1, SIDE 1

DR. TATAREWICZ: It seems these are briefing charts, as you indicated to us a moment ago, shown to Martin Collins in 4 1/2 inch slide form, and now you have them in xerox form, and you indicate that you'd like to talk a little bit about them. We will need to identify each one of them by its title before you start talking about it, so then we can correlate the charts along with the transcript of this interview.

DR. SEAMANS: All right. The first chart is a standard organization chart of NASA, as we discussed.

DR. MAUER: May I interrupt just for one second? It's the one titled "National Aeronautics and Space Administration, effective date November 1, 1963."

SEAMANS: It's one of the three or four that you got out of your file. Mr. Webb felt very strongly that the overall organization of NASA should be his responsibility, and any changes that were to be made would be made by him, and that no chart was official unless it was signed by him, and this particular one came after the long discussions that we had on how to continue with the Apollo program. Just prior to this, the organization was highly decentralized, and all of the centers, as well as the program offices, reported to me as the associate administrator.

    I only show this here this morning just to say that NASA was an organization with a variety of objectives, and it obviously had to have some kind of structure to tie together the Headquarters operations, on the one hand, and the field operations on the other. I'm not going to take any time here this morning to discuss any aspects of it, except to note that at the time we went through the reorganization which led to this chart, we decided that all of the field centers should report to one of the three program offices, which were Manned Space Flight, Space Science and Application, and the Advanced Research and Technology.

    The next chart I have, which is called "Location of Principal Launch Vehicle Facilities" Figure 5, doesn't show all of the field installations. It only shows those that were related to Manned Space Flight and in particular those that were involved in launch vehicles. It shows first of all the Headquarters. It shows the location of the Marshall Center. It shows the location of the Manned Space Flight Center in Houston, now called the Johnson Space Flight Center and it shows the Kennedy Space Center. The purpose of this chart was to show that we had to truck around or move around, transport some very large vehicles that were way too large to go by train or by truck or by plane, and the logical way was by boat. We worked closely with the Navy on this, and we got what was called roll on, roll off kind of barges.

    You can see in the chart that we had to move the first stage from the Michoud plant in New Orleans to the Mississippi Test Facility, and then around the tip of Florida to the Kennedy Center. The second stage was manufactured at Seal Beach in California. It had to go through the Panama Canal first for testing in Mississippi, and then to the Cape. The third stage was also made in California, but it was small enough so that it could be flown, first to the Mississippi Test Facility and then to Kennedy, but it was in a very unusual airplane called the Pregnant Guppie which was actually made from a conventional airplane by a very small company that was so short of cash that when they wanted to get the FAA to certificate the plane, they didn't have money for the gas, for the fuel to fly the airplane. The first stages, I should call them the prototypes, actually were made in Huntsville, Alabama. It's on a waterway system that permitted the transportation of the stages by barge on the Tennessee, Ohio and Mississippi Rivers, and to the Mississippi Test Facility and then around to the Cape.

MAUER: What does this chart date from?

SEAMANS: Well, this chart goes back to a very early part of the program, when actually, before President Kennedy had made the decision that we'd go to the moon with men, we had already carried out a very elaborate PERT kind of analysis of what was going to be involved. I think we had on the order of 10,000 different tasks that had to be carried out, and some could be done in parallel, some had to be done in series, but it was obvious from this analysis that the pacing item, what we call the long pole in the tent, was going to be the construction of the facilities that would be required. We needed very large facilities, and before you can build facilities you have to decide where they're going to be located.

    So very early in the program we had to make the difficult decisions related to the location of large facilities, which are obviously of interest to every state in the union. We had to have some rationale for making the decision, and this chart was made and ones like it, so we could explain to the Congress and the White House why it was that we were planning to use coastal sites or sites that could be reached by river, and also so that we could carry on discussions on why we wanted to have these sites in the South, and preferably in the Gulf area. We finally did make the decision that the second stage would be made in California, because North American, now Rockwell International, came in with the best plan for constructing and fabricating and designing that stage.

TATAREWICZ: Was there any concern about the Michoud plant? Was the Michoud plant essentially booked up, in terms of the use of the facility, such that you couldn't use the Michoud plant or expand it? I'm just curious about the route through the Panama Canal.

SEAMANS: Well, I think I went through this with Martin, but just let me say, one of the very early decisions had to do with the Michoud plant in New Orleans. Its availability was brought to our attention by von Braun and his people at Huntsville. This very large plant had roughly 40 acres under one roof, with full clearance throughout. It was a plant that was first used in World War II for shipbuilding and then in the Korean War, they made tank engines there. What we did was to manufacture the first stage of the Saturn I and the first stage of the Saturn V vehicles there, and that absolutely loaded the plant, and so there was not room to also manufacture the second stage of the Saturn V in this plant.

    Now, one possibility and one suggestion actually, made by North American, was that they build another plant somewhere in the Gulf area, possibly in Alabama or Mississippi. But we were not enthusiastic about that suggestion, it wasn't just a question of facility costs, it was also a question of manpower. We felt we were already saturated in the Gulf area with aerospace technology and a tremendous influx of people to the Cape and to Texas and Louisiana and Mississippi, and we thought that from a manpower standpoint, as well as from the standpoint of using existing facilities, we were better off to take the loss in time in transporting the 2nd stage around and through the Canal.

    I didn't bring along pictures of each one of the facilities involved, but the one I did bring was the Michoud assembly facility, New Orleans, Louisiana, with the figure 7 on it but that's a NASA number on it, MC 66-5451, and the date on it 1/25/66. Of course, what I brought in doesn't clearly tell but it does show the original plant, and the high rise area which we needed for static testing. It's not detailed enough to really get into a discussion of other parts of the facility which were added for engineering and other purposes, but you can clearly see the waterway going around it. The dikes were, I forget, ten feet high or something like that, because in very high water in storms and so on this whole area could become completely flooded if it weren't for the dikes.

    Having gone over all the organization, and the location of facilities for the Manned Program, let me double back and just say some general words about how decisions were made in NASA. The charts I happen to have are not all directly related to the Manned Flight. I have one here that's actually out of a paper that I wrote, just after I left NASA. The title of the paper was "Action and Reaction," and this particular slide is called "Research and Development Activity Leading to Proof of Concept." The purpose is to make clear the fact that there are a variety of factors involved in developing new ideas.

    At the start of any program, the decision has to be made to go ahead and do it. The decision-making has to be based on inputs that relate to need: why do you want to do it? External factors, internal factors, budgetary factors and so on must be considered. The decision-making for us on any of our major programs had to involve NASA itself. It had to involve the executive office and also had to involve the Congress. Out of the decision-making came resources, measured in terms of dollars, people, and facilities and so on, and then the implementation would be carried out in NASA by the program offices and the centers, and in turn all of the contractors that were tied into the organizations, through a wide variety of contracts. Out of this would ultimately come a new system, i.e., a launch vehicle or a satellite. The particular diagram shows the output to be a proof of concept meteorological satellite. That's what we needed to carry out the mission. The results would then feed back into the decision makers, so that they could decide whether to build another one or whether the system wasn't going to provide what was needed, was not going to satisfy the users, and whether it should be dropped.

    The next chart which is called "Balance Between Performance Schedule and Cost," shows three buckets instead of the usual two buckets in which you put your weights on the one side and the object to be measured on the other. Project managers have to weigh costs,time, and performance. It leads into, on the cost side, discussions of budgets and financial controls; on the time side, flight schedules, as well as a lot of other detailed schedules; and here's where computer analysis, PERT charts and so on are handy. Then on the performance side, you state what you want, the overall mission objectives, and then the Project Approval Documents spell out the performance that we wanted in more detail. Running these programs involved a continual balancing act between what you want to achieve on the one hand, how much it's going to cost, and when you hope to achieve it. When looking at the performance from the administrator's office, some way of scoring was needed, and that's when we got into an overall measure called Mission Success.

    I've got a chart here that shows the space flight record going from 1958 to 1965. We felt that to be meaningful we had to be hard boiled about how we measured mission success. We'd say, we either achieved the results that we wanted in our decision- making or we didn't. We didn't allow ourselves the option of saying that we achieved 58 percent or 93 percent of our objectives. It would then have gotten to be a job for a debating society to decide on the number. We'd just say, either the rockets fired properly and the satellite achieved orbit and we obtained good data--and on that basis, you find that in 1958, there were no successes. In 1959, there were 14 launches, and 60 percent of them were successful. Then we struggled along and in '60, there's 17 launches and we had maybe 55 percent. We then went on sort of a roll. We kept improving the results. In '61 we were a little over 60 percent. '62, we had 17 missions and we got up to 70 percent, and then in '63 we had 13 launches, with somewhat over 80 percent success. It levelled off there.

TATAREWICZ: So each mission then was either counted as a zero or a 1.

SEAMANS: That's right.

TATAREWICZ: Either it was successful or it was not, in the final judgment.

SEAMANS: Right. We did show this to the Congress, but it was very helpful internally as well, and we found that there could be a considerable argument about the score. If the particular project office had worked hard and placed a satellite in orbit but only transmitted data for a week, let's say, and then went off the air, they could claim, well, we got most of the data we wanted, it's pretty successful. I would have to note that it was supposed to operate for a year. That's what we agreed in the beginning, so it's a failure.

MAUER: Could I ask a question about that?

SEAMANS: Yes.

MAUER: One aspect of the investigations that has resulted from the loss of Challenger has been to show that there was perhaps a sense within parts of NASA, particularly of Marshall, of saying "as long as the vehicle gets launched and gets into space, then we've been successful." That type of perception, it seems to me, is very much what this procedure you've just described seeks to avoid.

SEAMANS: Well, sure. Now, it could be that prior to the Shuttle launching, that the performance that was discussed with Marshall, and maybe with that particular office, the Shuttle Office was solely to launch successfully and get the Shuttle into orbit, period. If that's what you'd agreed to beforehand, then Marshall would be right. That would be 100 percent. If that's not what was decided at the beginning, if the objective was to get the Shuttle into orbit and then to carry out a wide variety of experiments and return safely, and if orbital insertion is all that's achieved, then the mission is a failure. This is sort of you might say, too hard boiled, but you need some way of forcing a careful delineation of what the objectives are. If people know that the objectives are not there on the wall and it's hoped that we can achieve them all but if we don't, if we achieve most of them, that's satisfactory, then that's the kind of performance you're going to get.

MAUER: What were the circumstances that led to the formulation of this policy?

SEAMANS: Perhaps policy is a little strong. As general manager, I reviewed all these programs on a monthly basis, and I thought I needed some way of putting our achievements right there on the screen. After a satellite went into orbit, was the mission successful or not. I needed some way of talking to Mr. Webb and Dr. Dryden, the administrator and deputy, and going before Congress. Even though some could say that the chart I have here may not look as pretty as you'd like, it's only up to 80 percent, it certainly showed that there had been good progress during the period from '60 to '63.

MAUER: Was there also an internal aspect to this? Was it also a way of getting the various center and project management efforts to perform better?

SEAMANS: Absolutely, to perform better, and on top of that, to think clearly and precisely about what we're trying to accomplish before starting. People get so excited that they're going to the moon, putting up a synchronous satellite for communications or for weather forecasting, that they don't go through the agonizing job of defining what they're talking about. They're going to put a satellite at 800 miles and at a certain inclination angle and it's going to have three different kinds of cameras that take photographs in the infra-red, the ultraviolet and the visual, let's say, and really nail down what's supposed to be accomplished. If you don't do that, you can't talk about cost and schedule either. Everything's fuzzy. You have nothing to work against. You've got to be working against good hard standards, if you're going to budget properly or schedule properly and determine whether you succeed or not. I guess, you'd have to say this is my philosophy.

MAUER: Well, have there been problems that helped you to formulate this policy or this approach to dealing with things so clearly?

SEAMANS: Oh, yes. We've had Vanguard going through ten failures at the very beginning of the space program, where the public could see Vanguards blow up on live TV, and we blew up some Centaurs, and we were looking pretty sick in the world arena, for one thing. The confidence that the Congress and the public had in the program was not what we wanted. We were getting lacerated by the press. We had to be pretty hard boiled inside of NASA. That's what I was brought in to try to do as general manager. It's not always easy. When I was in DOD I ran into exactly the same situation. There was a tendency to say even though we lost X number of planes, the missions in Southeast Asia were a success. Doctors say it, you know, "the operation was a success but the patient died."

TATAREWICZ: So the Project Approval Document then was the forum in which the criteria for success were spelled out and then agreed to by all concerned?

SEAMANS: Exactly. Before you start. Then you either do it or you don't.

MAUER: So the Project Approval Document becomes something analogous to a contract.

SEAMANS: Exactly. That's exactly the idea. And Webb used those terms. He would say, "I contracted with Bob Seamans as general manager to carry out these missions, and he in turn subcontracted to the program offices."

    Here is another one of these success charts. This is for the year '65, which turned out to be 82 percent. The 1965 record of mission success and percent to December 30, '65, the very end of the year. Along the bottom, you can take a look at, there were 28 missions. It says, success 23, failure 5, success ratio 82. It also breaks out launch vehicles, of which it says there were 30, and I don't quite know why the 30 and the 28 aren't the same. I guess we had a couple of vehicle launches that weren't related to missions, then they're just launch vehicle tests. Perhaps the Saturn I fits in here. I'll have to see as we go through it, whether it's 26 successes, for an 87 percent success rate, and this is where Gemini 3 was involved first. That was on January 19th.

    You can go through a long list of these flights, and it looks like one, two, three, four, five were successful. So when the Centaur failed on the sixth, it dropped the average down, if I'm reading this correctly, to around 85 percent, then it built up, and there was another failure and it dropped down--that was the Little Joe test, that's one of the vehicle tests. Then it built up again, and then we had three failures in a row, and then it built up again to give us the 82 percent overall. At the administrator's level, you've got to have some fairly simple straightforward measures of how everybody's doing, and I'll get into a few more of these measures, with these particular charts.

    Just one thought as it comes to mind, though, on this success business. Every time the Shuttle was launched, for whatever purpose, there were people aboard, and so mission success obviously involves the success of all the payloads that were in the Shuttle Bay. Overall it also involves having that Shuttle take off successfully, go into orbit and land successfully, not only with all the parts together but with the crew still in good shape. When you ask, looking at the past, what could we expect in the way of performance?

    Now, in the seventies the performance figures were coming in higher than this 80 percent. We've never seen any overall that came in over 95 percent. You can say, okay, we're going to get one failure every 20 or so. We had 24 successful Shuttle launches and then one failure. You can see how catastrophic that failure was. Yet when you look at the statistics, it wasn't really surprising, if you look at it that way. If you look at what information NASA had to deal with, however, prior to that failure, you'd say that they should have done something about it, that there were indications that they had scorching and burn through of the O rings in 15 of those 24 flights.

    Last year when we carried out a study of the Space Station for the executive office, we got into the question of, what's it going to take to put the Space Station in orbit? The answer is, to put the complete Space Station in orbit, it's going to take 33 launches of the Shuttle. Now, we're trying to make one of each kind of module. You have to have all these modules working in orbit to end up with a complete Space Station. What's the chance that you might need some backup? Well, you can see that the odds were against all of them succeeding, and we strongly recommended backup hardware for the Space Station, and we strongly recommended that ways be found to improve the reliability, performance and safety of the Shuttle. It's really based on this kind of thinking.

    Now, I don't say this is the most orderly presentation of management we have here, but these are the charts I've already shown Martin. I have one that I did discuss with Martin, one on unit use, the Management Organization for the Apollo Program, and it shows on it, up at the top, the Apollo Program Office, Washington, DC, and this was the manager of Apollo--Sam Phillips. At that time he was on assignment from the Air Force, he was a lieutenant general. After he left NASA he became a four star general in charge of the systems command of the Air Force.

    This method of management was really put into being by George Mueller. He took Brainerd Holmes' place, and we discussed that transfer at some length. He came in, and the idea is that you have the three centers, Houston for the manned side of it, Huntsville for the launch vehicle side, and Kennedy for the actual launch facilities. That's not to say there weren't other parts of NASA involved, but these were the principals, and within each one of these centers would be the responsibility for some element of Apollo.

    The question was, how are you going to put this whole operation together? We'd already been through several unsuccessful exercises in NASA where we came to the conclusion that we couldn't have a straight line organization from the head of a program office, or the associate administrator, to a center director, and then down to the particular projects, that it was just not humanly possible to funnel everything, all the information up and down this line that way. So this shows how it's broken apart.

    Just to take one example, the Apollo Spacecraft Project Office at Houston, and the individual responsible there would really wear two hats. With one hat on, he was reporting administratively to the director, Bob Gilruth, in that particular case, and up to the program office. With the other hat, This chart doesn't completely show all the lines--he would be reporting directly to Sam Phillips, for his project direction, and he would have reporting to him the same kinds of functional offices that reported to Sam Phillips.

    These offices were the program control, this is really the fiscal control, comptroller; the systems engineering, that is, keeping track of what it is that's being designed, all the elements that make up in this case the spacecraft, with the specs and so on for each one; the test procedures and planning that were to be carried out on the particular elements; the reliability and quality assessment; and finally the actual flight operations, where the spacecraft was going to be taken to the Cape and integrated with a lot of other things and then shot up into orbit.

    Similarly at Huntsville, you have the Saturn Project Office, and the Saturn V Project Office, each with its own functional offices reporting, and in turn the same kind of arrangement down at the Cape, where there was a Launch Operations Office responsible for building all the facilities. Rocco Petrone was in charge of the project at the Cape reporting administratively to Curt Debus, who was the center director but he reported to Sam Phillips from a program standpoint. These functional people also reported directly back and forth with each other.

TATAREWICZ: Was there any opposition to such a thing from the center directors?

SEAMANS: Tremendous. But they went with it. I already have discussed with Martin though how we got into this kind of thing at Goddard, way before we got into it in manned flight. Harry Goett, who was a personal friend of mine I'd known for years, refused to go along with it. He said, "I refuse to have program people from Headquarters prowling around my center talking to my people, telling them to do things. How can I run the center when I don't know what's going on?" He said, "Everybody who comes out to the Goddard Space Flight Center is going to see me first, and I'll tell them whether they can see my staff and project people or not. And I'll sit with them when that takes place."

    There were something like 50 projects going on at Goddard. Harry Goett didn't have time to do that. It finally reached the point where we told Harry, if he couldn't cooperate the way we wanted, he'd have to go and find another job. That's exactly what happened. These programs were so big and so complex that you couldn't operate the way the old NACA had formerly operated. At the same time, these project people, say the spacecraft at Houston, obviously have responsibility to keep Bob Gilruth informed of anything unusual that happened, particularly anything that would involve additional resources, either in terms of space or dollars or people.

    On top of that, though--and this was the really innovative part--George Mueller as head of the whole Manned Flight Program and director of the three centers would have his directors' meeting with him once a month. He would sit with Wernher von Braun and Curt Davis and Bob Gilruth and then Sam Phillips would come in and discuss the Apollo program, just as Chuck Matthews would come in and discuss the Gemini program, and somebody else would come in and discuss the advanced Apollo program. There's no perfect way to organize of course.

MAUER: That's clear. It's always a matter of making compromises. But what about the alternative approach, in terms of having a lead center, instead of having a program office at Headquarters, giving one center the responsibility for managing a particular program such as occurred with Shuttle and JSC?

SEAMANS: That's of course what NASA came to, over time, on the Shuttle program--for the human reasons we're all aware of. The centers didn't really like to have Headquarters people watching them, telling them what to do--I'd say that's why NASA went to the lead center concept. The lead center concept is a good concept for programs where, 75, or 85 percent of the program or something on that order can really be carried out by a particular center, and they'll call on other centers for 5, 10, 15, 20 percent.

    But in the case of the Shuttle, clearly NASA didn't have those kinds of percentages. I don't know what the numbers are, but probably Houston had about a third and Huntsville a third, maybe a third at the Cape or something like that, and the system fell apart. On paper, at the time of the accident, the program responsibility for those solid motors, went from Marshall to Houston and then in here to Headquarters, through the associate administrator and administrator.

    Among other problems, Houston didn't have the clout in Houston to force Marshall's hand, and Headquarters didn't have a full understanding of what was going on. When Jim Fletcher came in as the administrator for NASA, he got Sam Phillips to review NASA management, and I would say that the foremost and most important recommendation Sam Phillips made was to set up a program office for the Shuttle, which is now physically located out at Reston. This was obviously not enthusiastically received by the centers. I guess that's another thing you can say, that over time these things do tend to ebb and flow somewhat, because these are real pressure points.

MAUER: I understand from one source that there's a feeling that there were problems, some people in NASA had a sense that there were problems of over-centralization. You had the program office for Apollo located at Headquarters, and that was one of the forces that led to the lead center. Is that your understanding?

SEAMANS: I'm sure that would be said, all right. Of course I left NASA in '68 and I followed it from afar, and I think as you get into this oral history and you bring in the people who followed Mr. Webb and myself and so on, you'll be able to better understand how things change with time.

TAPE 1, SIDE 2

SEAMANS: Let me just say, repeat, that there isn't any perfect way to run these programs. I think one of the really important things that the Smithsonian can do with this oral history is to get at just this kind of an issue, because it's very, very fundamental, for anything that you do on a large scale in this country in the future. Hopefully there are some lessons to be learned by reviewing Apollo and subsequent programs, and of course a simple-minded view of this by someone who happened to be there in the sixties is that we tended to prove the centralized method could be made to work, and the decentralized lead center in one important situation did run into trouble. Perhaps that's too simplistic. I think it's important to really get into more depth, and see why the one didn't succeed in a particular case, and the other did succeed, and in part, it's not just how you line up the diagram, it's who the people were that were involved.

MAUER: In your experience while you were at NASA, how did you feel about this management structure that was used in Apollo?

SEAMANS: Well, of course I was party to it and I helped to put it together this way, so I guess I'm prejudiced in its favor. I certainly understand though that you can overly centralize, and I really tried not to get myself into too much detail, but to only get into detail when there seemed to be some kind of incipient problem or we'd had a problem, in order to be sure that we had the right resources and the right teams of people to square things away. Then to back out of it, and not to micromanage, a term that's used very much today, it refers to the Congress of course getting into too much detail. But it's one of the most fundamental issues, I think, that we face, from the management standpoint, in this country. There's a terrible danger, as we run into problems, that the statement is made "Well, the answer is, establish another department, have another member of the Cabinet, have another administrator, associate administrator in NASA reporting to the administrator."

    I'm very concerned NASA has been forced to establish a quality control section. There is now a whole quality control office that's independent of the program arrangement that I'm describing here, where the head quality control official at NASA could be sitting at the Cape before every Shuttle launching with the power to stop the launch. I think that's ridiculous. I think, I'm for quality control if it's reasonably embedded into the program operation as was the case in Apollo. But to have a completely separate group of people with absolutely nothing to gain by going ahead with the launch, really, and with everything to lose if the launch fails, sitting there as one of the finalists sitting around with the launch operations team, saying Go or No Go.

MAUER: How do you contrast that with how you handled quality control in the Apollo program?

SEAMANS: Well, in this particular chart here, we had the quality control in the program offices. We did not have a separate quality control office. When I arrived in NASA, we had a reliability and quality control. It was headed by somebody named Nick Golovin, and what he was pressing on was a rather simplistic method of measuring quality. It had to do with putting the reliability figures on each part, and then by noting how the launch vehicle is assembled and operated, it's possible to determine how the overall reliability hinges on each one of these parts. In some cases there is redundancy, in some cases there isn't, and so on.

    You can make a reliability tree, if you will, for the overall system. This arithmetic generates a number, like 70 percent so say it's then possible to--there's a 30 percent chance that this thing's going to fail. These kinds of numbers are very, very dangerous, in the sense that they can be used out of context. They should be used and were used to indicate how to put a system together to maximize the chances for success. Two ways of configuring a system can be compared. One can give you the best chance of success. Another may put the astronauts in greater jeopardy from the standpoint of being able to recover them if there's a failure. But if these numbers are really believed and they start floating around, there can be trouble. Nick Golovin was still there when we were looking at the large launch vehicles like the Saturn V, and I think the numbers came up, on every launch we had a 50 percent chance of failure, but as a matter of fact, the Saturn V worked every time, defied the numbers you might say.

    Reliability and quality control require much greater understanding than that to determine where very tight tolerances are required and where not. The inspection procedures must be appropriate to the demands of the system. That's the kind of thing that reliability and quality assessment were involved in, to be sure that there were proper specifications for the tolerances of the critical parts, proper inspection procedures and so on. You probably know that we had the astronauts visit the plants where the critical parts were being made, where they met and addressed the people working there. When you finally get down to it, it's not a question of automatic machinery getting the quality, it's a question of people, and motivation and so on that's very important.

    One last thing on this chart before we leave it, in a way it's a shame, it's out of balance. Ninety percent of the chart has to do with NASA and its centers, and about 10 percent of the area shows contractors, at the very bottom. And actually, when we really got Apollo going at its peak, I think only 5 percent of the people involved were at NASA, and the remaining 95 percent were in US industry, with a very few percentage, maybe one or two percent in universities. This was a conscious decision that was made early on, that we would increase the government side, namely the NASA side, a minimal amount. I guess we did nearly double the size. We finally got up to something like 35,000 people. That's the total number of people working on the program including Apollo, I think, there were on the order of 450,000 people on contract.

    But before the contractors could work, there had to be a procurement. First we had to select the contractors, and then we had to have a way of determining whether the contractors were doing their job or not, and that gets into this next chart, which I discussed quite extensively with Martin, and I used this very chart. It was on a glass slide. This says, "Procedures and Authority for Procurement Actions," and across the top, there is the first project manager at the field center and then the center director and his staff and then the Headquarters program office, the Headquarters staff and finally the administrator.

    Then down the side, the first line starts out with the procurement plan, and next there is the Request for Proposal, the RFQ, and then the evaluation by the Source Board, the selection of the contractor, and finally a contract is signed between the government and the particular company. This shows that the procurement plan is initiated at the project level, and is reviewed by, among others the center director, the program office at Headquarters. We did have reporting to me the procurement officer, who's very familiar with all the latest regulations and so on, partly from the Congress and partly implemented through OMB.

    On the large procurements--which I think were those over five or ten million--the administrator gave the final approval, and I actually approved the RFQ, and then Mr. Webb took great stock in the evaluation process, and spent himself, he and Dryden as well as myself, a lot of time on what the Source Evaluation Board came up with. We met with the SEB. On the big ones we'd sit for maybe a full day, asking lots and lots of questions, and then before the final decision was made, before the selection was made by the administrator, we would invite in the center directors and the head of the program (Sam Phillips in the case of Apollo) and we'd say to them, okay, the Source Evaluation Board has come in with their findings, and they would not come in and say you ought to pick a certain company. They would evaluate the companies with respect to one another, and we would ask Sam, "okay, is there anything that you know of, in addition to what we've heard this morning, that we should know about before making the decision?"

    In one very important case, namely the selection of North American, now Rockwell, for the spacecraft, what was said at that time led to the selection of North American rather than the Martin Company. Bob Gilruth said he was very concerned that the Martin Company hadn't made any airplanes in, I forget how long, ten or fifteen years, whereas North American, who came in second, was very much in the aircraft business and had been responsible for the X-15 which was the closest thing we really had to what we were going to build in Apollo.

    Then, having fed that in, everybody would leave the room except for Webb, Dryden and myself and whoever the executive secretary was for Mr. Webb, and the three of us would discuss it. Usually being the junior person, I was asked first what my recommendations were, then Hugh, then Jim Webb would actually make the decision right there with us. When I gave my recommendation, I couldn't just say "Pick so and so," I had to give the reasons for it. Then the executive secretary, when it was all over, would put all of the points in a finding of about two pages, with the reasons, and we all three would sign it.

    One of the satisfactions of the job was that we never had one of these actions overturned, which was fairly unusual, because any contractor who wanted could go to the GAO and ask for a review, and at that point, the procurement can't go ahead until the GAO completes its review. In any event, that was the process that we followed. After the announcement has been made, that negotiations are going to be initiated say with Rockwell, work can proceed much too long on a letter contract. I haven't shown it here, but one of the charts that I kept was the amount in dollars that we had on letter contract. We tried to crank that number down, at the earliest possible time in order to achieve a definitive contract.

    A fellow named Bob Charles was brought to NASA by Mr. Webb. He knew Bob Charles at McDonnell Douglas. I guess Mr. Webb had been a director there, and he brought in Bob to look at the kinds of contracts that we had, and Bob worked with us for about a year before going over and becoming Assistant Secretary of the Air Force for procurement, and on the basis of what Bob did, and I guess our own feelings on the matter, we felt that the more we could get on incentive contracts, the better off we'd be. We recognized that you can overdo incentive arrangements. You can make them too mechanical, particularly in development, you've got to be careful that you don't over-emphasize costs, you can keep the costs down but you end up with something that doesn't perform.

    This particular chart shows the extent of NASA and center contracting in billions of dollars, and the date on this is 3/15/66, and the time the chart was drawn, we had 159 contracts, an accumulated dollar amount of over two billion. I don't really remember, when we finally finished Apollo, how large that number gets, but I'm pretty sure that we ended up with all of the major contracts on incentive arrangements.

    These incentive arrangements ran all the way from an award fee, where it was really that the award would be based on the judgment of the government, to something that was fairly mechanical, just based on dollars. You know, we could talk about this ad nauseam. But even in the case of the award fee, we felt that it shouldn't just be the government deciding that the company did a pretty good job. It ought to be based on a series of reasons, and the reasons ought to be stated before the contract is signed. Then each time that a decision on the award came up, the decision should show which factors the contractor scored well on, which they didn't, and that in turn was discussed with the contractor. The reason for this exercise was the discipline required on the part of both the government and the contractor. Each had to decide ahead of time what they were trying to accomplish, and then the people carrying out the work were judged on the basis of these predetermined objectives.

    This chart really doesn't relate directly to procurement, but I did discuss this with Martin. There have to be a series of milestones, in any of these programs, although in most of them, at the Headquarters level, we emphasized the launch and we kept track of the launch date. The program people themselves had to have a whole bunch of interim milestones, that they had to pass, and to the extent possible, we put in the procurement phases, phases of procurement in with these milestones, and what this chart shows is the Apollo review process. The figure 15's on it, there's no date on it, but it shows the definition stage to begin with, when preliminary design takes place and is reviewed, and then there may be a procurement process here, since several companies may be involved in the definition phase. There can be several contractors all the way through in some cases.

    In the critical design review, there is usually a complete mockup. All the elements in the mockup can be inspected, and the details of the specs may be discussed. Then the first article is fabricated, it may not be for flight but at least it's going to be tested say at Mississippi Test Facility. After the first article configuration is inspected, the next phase is prototype testing. Then the operational side, there is the design certification review, when all the specs are really nailed down. From then on an attempt is made to minimize change. In the case of Apollo, we set up a board so that any change had to go to this board for approval. The board was chaired by Sam Phillips.

TATAREWICZ: Configuration control?

SEAMANS: Configuration Control Board, yes, and at the time that the decision had to be made on something like that, the dollars and the schedules and so on would also be brought up, and finally, just before the flight, maybe a couple of weeks before a flight, there'd be a flight readiness review. I sat in on a few of them, some of the important ones, sort of in the background; I wanted to see how they went. I hate to think how many people were there for some of these Apollo launches. There may have been 250 people in that room.

    The final charts, I have--they're pretty simplistic, but again, we get into this whole matter of performance and time and dollars. The first one shows across, this is now called "The Project Manpower Versus Time Original Estimate." Across the bottom, time--you again have the study phase; the preliminary design breadboarding and design fabrication; ground test, the next phase; and finally the operational phase where we achieve launch. Then we get operational and experimental data and so on coming back. In general, these things build up to a peak somewhere during the design fabrication phase and then tail off.

    It ought to be made clear that when you go from say design and breadboarding to the fabrication and ground test, it shows what is happening at one instant of time, whereas you may get part way into design and fabrication and run into a problem and at that point we would have to go back and redesign and rebreadboard it. There's a lot of back and forth there. But the purpose of the chart is to say, okay, that's the first estimate, but that's not necessarily what is going to happen. There may be delays for God knows what reason. But because of the delay, and the extension in time, an important point of this chart is to show, this is now a comparison of actual and estimated manpower, and this had a NASA number. I don't know, it's 3/15/66, and you can see what happens when you have these delays. There must be manpower around for longer than originally estimated, and the cost of the program is going to go up, depending on the difference between the integral of those two curves.

MAUER: Right. Can I ask a detail that you probably won't remember, but it's worth asking anyway. A fair number of these charts have been 3/15/66, and was there something that was happening at that time that led to briefing being prepared?

SEAMANS: I guess the first Goddard lecture, and I think I had some of these put together for that lecture.

MAUER: Okay, good.

SEAMANS: Yes. When I joined NASA, let's see, I was the second general manager or associate administrator. Keith Glennan early on had decided that neither he nor Hugh could really be the internal manager of NASA. They both had plenty to do with all the external affairs. Keith brought in Dick Horner who had been the Assistant Secretary of the Air Force for R&D and he went to work for NASA. He was a tremendous help to me in getting started, and by the time I got there, there was one room with nothing but charts all around it, and these were sliding charts, pretty good sized, something like, I don't know, 30 by 40 inches.

    On the charts had the various milestones, launch dates for each project. The milestone to be achieved at that instant could be shown by diamonds pointed upward that would be black, filled in, when the milestone had been achieved. They would be not filled in when they were the goals that you were working towards, and they would be dotted with dates on them for the ones that were the goals let's say the previous month or the month before that. After a while they could get complicated. It's very hard, looking at them, to get much of a sense of what's going on. So I suggested, rather than show them that way, why not take a few critical items, like a launch date, and just plot the planned date versus time.

    That leads to this chart, which shows schedule history, and this isn't a particular project, it's a typical trend chart. Across the bottom is the actual time, real time, and then vertically is the time when the launch or some other critical date is anticipated. In an ideal world, we would hold that date, it wouldn't slip or slide, and by God, when you got to that date, the launch would take place. Ideally, if you drew a 45 degree line, that would tell you when anything was supposed to happen. You could say, we're going to do something critical in January '66, and follow right along, and it would line up on that curve. Very simple-minded. But it wasn't so simple-minded when you started. Project managers started projecting what they were going to do. The tendency was to let things slide for quite a while, because project teams hate to admit that they've got a problem. But the truth has to come out, at the very latest, on the date when a goal is supposed to be reached if it doesn't. and so this becomes a fictitious curve. It makes the point that you tend to go along holding that date, and all of a sudden there's a big slip, and we kept charts like this for every project. I'll have to say that few liked them as well as I did.

MAUER: How did you make use of that?

SEAMANS: I met with the project people every month. No, that's not true. I met with the program people every month. Now, I wouldn't necessarily flash on the screen every one of these trend charts for all the 70 different projects we had under way, but whenever there was a problem I could trot them out. But they were always available. I've never been able to prove this, but in some cases--and I'll give you an example of one--maybe it was too easy a project or maybe it was just the excellence of the team, they came pretty close to the ideal. In other cases, once delays started they tended to keep going and going, and yes, there were a few cases where these curves were divergent. We could show that after involvement in the program for two years, the goal was further away than when the project started. They're quite revealing. I would say, this is not how you run a program. You've got to run a program with charts that look like this.

    I've just picked out two particular Gemini launches. One is the Gemini 7 and the other is Gemini 6, and this sometimes called Gemini 76. The idea was to put Gemini 7 up first, and then--and this is very, very tough--within less than two weeks, Gemini 7 could only stay in orbit two weeks, we'd get off Number 6, and show that we could rendezvous, obviously we couldn't dock, but we could at least fly in formation for a while. We only had one launch pad, and for a long time we didn't think we could operate on centers of less than a couple of months, so the turnaround on the pad of a week was pretty unique, and the scheduling for that week, it was actually a little more than that, it was ten days, this showed the various things that we had to do after we launched 7, called the Gemini 6 Launch Operation.

    Just a few of the items on here, we had to erect the next Titan, we had a whole set of different tests and so on that we had to run, in order to get off as the schedule was on the 10th of December. Take it back, the launching was on the 4th for Gemini 7, so the original plan was to get off on the 10th, which was six days from the time that 7 was launched, and as I remember it, we finally got it off on the 12th, which was eight days afterwards. But then, even this detail wasn't enough for the people down at the Cape to know what they're supposed to do, and that's the reason that had this chart put together, to show what had to be done down there on the pad on one particular day. It probably says somewhere on here which day it was. But it shows all the detailed things that took place, and it's scheduled throughout the complete 24 hours, of what had to be going on in order to achieve the desired result.

    Then finally, the final chart I brought in, was to show the best project from the standpoint, as measured by holding dates, of any program that we've had at NASA. This was Chuck Matthews' project. This is the Manned Space Flight Schedule Trend chart for the whole Gemini program, going from 1962 all the way to the end of '66. In the very beginning, and this actually really started in '61, when we were really starting to get serious about going to the moon, the people at McDonnell-Douglas and the Mercury team said, "You know, you don't need to go, you don't need to have a great big launch vehicle. You can have a two man team fly to the moon by making use of the Titan vehicle and modifying the Mercury by making it larger." At that time it wasn't called Gemini. Most people really didn't take that proposal very seriously. They thought they were going to have to go to a whole new launch vehicle and so on. It had to be a three man operation.

    But there got to be quite a debate over this, and some people in the White House actually thought we could go with a two man operation, and they pointed out we were only going to have two men land on the moon anyway. But by the time we got into Apollo, to the tune of I guess about half a year, and it was the start of '62, we'd pretty well decided that we were going to have to get some operational experience in a hurry, and the operational experience was maneuvering in orbit, including docking, and extravehicular activity. It was going to be a long time before we had Saturn and Apollo, and we thought there was a tremendous advantage in getting into this phase of the program sooner. So what had been called the advanced Mercury, the original name, came to be called Gemini, and was introduced as part of the Manned Lunar Landing Program.

    The original scheduling, the planning, was to have the completion of all twelve flights by mid-'65. But this was before we had negotiated the definitized contract with McDonnell-Douglas. We finally got the definitized contract worked out, it was roughly in March or April of '63, and it was for twelve launches of which ten were manned. The first manned was Gemini 3, and if you look at the chart you see that there was a slip of about, oh, something like a half a year for Gemini 1. The first manned launching, Gemini 3, slipped quite a bit. I guess that's not quite a year. But George Mueller, who ran the whole Manned Flight Program, and the project people and the Air Force, who were responsible for the launch vehicle, really did a miraculous job, and as you can see, the twelfth Gemini actually was launched sooner than we had scheduled as part of the definitized contract. There's no question that Gemini taught us a tremendous amount.

    One of the really amazing flights was the Neil Armstrong flight, I guess you could say it's because of Gemini 8, because of what Neil did that he finally ended up going to the moon on Apollo 11. But on that flight we linked up with an Agena capsule for the first time, and used the Agena motor to further propel the Gemini out of circular orbit. Neil was flying over China in Gemini 8 when he noticed that the whole kluge was starting to spin, and he figured the problem was probably in the Agena, so he disconnected, and then he really started to spin, because the problem resulted from one of the thrusters aboard the Gemini that was firing when it shouldn't, and before he could turn it off, he was spinning at a rate of two complete revolutions a second, if you can imagine doing that. I can't.

MAUER: I can't either. That's where the test pilot skills come in.

SEAMANS: That's why you want to have a pro at the controls. It was just so violent that even a very small motion of his head would change the acceleration of Coriolis, right here, where his vestibular is, and he got completely nauseated. But it had to be one of the thrusters, so he just turned them all off first. And then he had to de-spin, and he couldn't use his conventional system. The only thing he had available was the pack that was in there for control during re-entry, and just fortunately he had enough energy in that system to de-spin, and still had a little bit of fuel left for the re-entry. But it was a really close call. And when he was fighting this battle, he was completely out of touch. On the Gemini, we didn't have the communications that we finally ended up with on Apollo. Just to digress a little bit--are you going to change the tape?

MAUER: In just a moment.

SEAMANS: All right. It just happened that that was the night of the Goddard banquet. We had Hubert Horatio Humphrey, who was then head of the Space Council, as our speaker. When I left home for the banquet, Neil had docked successfully with Agena, so I was really very pleased, and I got in the car and went over to where the banquet was, the Shoreham perhaps, and then by the time I got there, I was greeted by somebody from NASA saying that we've got a very, very serious problem. And not a lot was then known about it. I got there just before going in to dinner, and I remember feeling that the people there at the dinner ought to know about the situation, because otherwise there would be rumors spreading all around. There were enough people in the room who already knew about it.

    So I remember standing up and announcing it before well over 500 people, that we had a serious problem, and people thought I was joking at first. They couldn't believe it. But needless to say, it made the dinner a little more somber than planned, and of course the press were on this and the news people, and Cronkite was on the air, and he was cautiously announcing what was going on, and trying to get information as fast as he could. And the information was such that we believed that we would recover Neil before Humphrey finished speaking. And I had a hand shake, I was sitting right next to him, and a hand shake that I would tell him just the second we knew, and he started talking, and I was gathering information, and he'd look at me, and then he'd go on for five minutes. You know, even Hubert was having trouble. So finally I gave him the word, and he announced that Neil Armstrong was successfully recovered, and that ended the evening on a happy note.

MAUER: That must have been a very stressful meeting for you.

SEAMANS: That was the meeting that I learned that I'd made a big mistake. The second I stepped out of that car and found out what was really going on, I should have gotten back in the car and gone to Headquarters. When we had the Apollo fire, we'd all been at the White House to witness the signing of a treaty. The President signed a space treaty with the Russians. I went directly home, and as I walked in the door, my wife said, "George Low is on the phone," so I went to the phone and George said, "They're dead." And I said, "Who?" And George could hardly speak. It turned out that we'd lost Gus Grissom, Ed White and Roger Chafee in the fire. We were having a dinner party that night, and I told my wife, have the dinner party, don't tell anybody what's going on, as long as you can keep it quiet, so they can at least eat some good food, and I headed back for the office. Boy, was I glad I did. I had very strange things happen. I was talking to McNamara, explaining what was going on.

TAPE 2, SIDE 1

MAUER: You were talking about McNamara at the time of the 204 fire.

SEAMANS: Yes. I'm not sure I was actually talking to him. I was talking to his exec. But the operator suddenly cut in and said, "Emergency call, emergency call," and it was one of the networks, Peter Hackett of NBC, and he said, "Bob, this is a real crisis. We know we've lost the astronauts, and the country is very unsettled by this, and you'd better come over right now to the NBC studios and reassure everybody." I said, "Peter, you realize I was having a conversation of some importance, and you just cut in on me, and there's no way I can come over to NBC and reassure the country. We don't really know what's happened, and we're going to have to gather information carefully before we give any kind of a press release or go on live TV."

    I couldn't believe that NBC had been able to bust right into the phone call I was having. I'd been talking to others. I'd talked to the Vice President about it and so on. But anyway, the real point is that that evening, in my office, where I had some privacy, I was able to make sure that everybody was informed who should be informed. I was able to get some reassurance from Houston. I talked to Dr. Berry, who was the astronauts' physician, and got into issues that related to the astronauts and their families. I had discussions with others about the nature of the fire, and how and when it had occurred. And the most important thing I got to do was to put into play a plan for a careful review.

    As a result of the Gemini near disaster, we'd gone back and taken a look at NASA procedures for any serious accidents. We had a number where we'd lost pilots. We lost a couple of our astronauts who were making a landing in one of the NASA jets and they were hit by a couple of birds, broke the windshield and we lost them. We lost a couple of other astronauts coming into St. Louis. They were going to McDonnell-Douglas, and they broke through an overcast, and were trying to stay under it and turn around and land and they hit a building. But those were manageable accidents, in the sense that they didn't really excite the press and the public. But here we had two national heroes, Gus Grissom and Ed White, who'd been killed in a catastrophic accident, and fortunately we had changed our procedures. There were still procedures that involved setting up an accident review board reporting to the associate administrator, and the one change I made was to have the board report to the administrator. Now with real 20/20 hindsight you can see that we might have had them report to the President as was the case after the Challenger disaster.

MAUER: A Presidential commission.

SEAMANS: But I talked with George Mueller, and I think by about 2 in the morning we'd agreed on the makeup of the board and agreed that Tommy Thompson, the director of Langley, who was greatly respected throughout NASA, would be its chairman, and that I would fly down at 6 in the morning and pick him up at Langley, and then we'd go down to the Cape. Of course, one of the first things we did was to impound all the data and other evidence. I checked this out with Jim Webb somewhere around 1 o'clock in the morning, and that's how we got the whole review board in operation. Mr. Webb in turn had to get some kind of an agreement both with the President and the Congress as to how we were going to deal at that level, and then in turn how we would report through the media to the public.

    The agreement he got was that as soon as we had information, significant information, we would take it to the President, and then within two hours it would go to the appropriate committees, and only after that could anything be released of a substantive nature to the press. However, these would not be reports from the review board, which would not have to report until they had completed their analysis, but rather from me. I would go down to the Cape on a weekly basis and meet with the review board, and then fly back to Washington writing a report en route. I submitted it to Mr. Webb who in turn would then carry the report to the President. That's what we did, I think five times in a row.

    The first report we wrote was actually printed in its entirety by the New York Times. We could have had a more complete procedure, I guess, but it's pretty hard to anticipate things, exactly how you're going to deal with such traumatic events. At the time of the Challenger accident, I don't know where you all were, I happened to be in a bank, and I heard somebody--it was a fairly good sized bank in Boston--say something about accident, and I knew Challenger was about to be launched. I went over to another desk, a vice president's. I said, "I hear the word accident." He said, "Yes, the Challenger was burned up or exploded." And then I heard that afternoon that Graham, who was the acting administrator, had gone with the Vice President to the Cape, and I had a sinking feeling that NASA was not going to be able to carry out the review.

MAUER: Why a sinking feeling?

SEAMANS: Well, because I think the Rogers Committee did an excellent job, I must say that, and I think the report they wrote was superb. I spent a lot of time on it, gave lectures on it. But I think that it's better if you've got the qualified people to do the job, to do it within the particular department or agency, bringing in a few people from the outside if required. We brought in an Air Force colonel who had been responsible for the inquiry into a Titan explosion in a silo, and we brought in the head of the, what was it called, the Bureau of Mines, the group that investigates mining explosions when a mine blows up.

MAUER: Mine Safety Board?

SEAMANS: Yes, something like that. That was invaluable. I think we were able to get this, the whole business behind us in a proper way much faster than was actually possible with the Challenger Review Board, where, because it was a Presidential board, they were actually issuing press releases as they went along. I think it's much better to have a board that can get the whole job done, and then treat it in its entirety, as rapidly as possible. Now, poor old NASA, I hope they're going to get off a flight this summer, but you realize that over two years have gone by already. And we got Apollo back on track I'd say in a matter of about three months, and all the changes and so on were agreed to and the whole thing redesigned in a matter of six months, I believe.

MAUER: Based on your experience with the Apollo Accident Review Board, and your experience working with being an observer of the Rogers Commission, how would you compare these two processes?

SEAMANS: I'm getting into an area I'm a little hyper on. Because I got so involved in the manned station, Space Station, last summer, I made a much more careful study of NASA, and I went through the various centers and so on, to see what's happened to NASA in the last fifteen or twenty years. There are a couple of excellent reports that have been written. The Rogers Commission Report is really amazingly good. And then there's a study that the NRC [National Research Council] carried out for--I forget who it was, did NASA ask for this themselves or not--anyway, it was on the whole matter of research in NASA; I guess they called it advanced technology or technological development. Joe Shea was the chairman of the study which was finished the end of last summer.

    What it showed was that, in the period from '65 to roughly '75, the amount of advanced research and technology effort in NASA in constant dollars dropped by a factor of about four times. Once it dropped it stayed at that level, until about a year and a half ago, and since then, additional funds have been put in their budget for special centers of excellence and things of that sort.

    But if you go back in time to the old NACA, there has been a lot of very creative research and study going on within the NACA that was self-motivated, self-regulated, you might say, but where the NACA was in close communication with the aerospace industry, and they had the wind tunnels and so on, and the people they had didn't earn as much money. They had, call it the fun or the excitement of having the best facilities in the world to work with.

    So you really had some, NASA had some, very able people. Then when the Huntsville team and the JPL team were married in, it increased the capability of NASA to really do advanced work and have people that, you know, really understood fractures. Mechanics, for example, were way out on the front edge of what was going on. So when you had a serious problem on the Saturn when we were hydrostatically testing the second stage and the damned thing came apart. It just unravelled. There was a hairline fracture, and at cryogenic temperatures, when the pressure reaches a certain level, the tank just unzippered, but there was enough expertise in various places in NASA that it was dealt with, and then we could move ahead. Whereas when NASA had these problems with the scorching on the shuttle O rings, they really didn't have the people to turn to, to deal with it and make sure that the right kind of redesign was carried out, and they didn't have an alarm system that triggered a stop order and a redesign. NASA didn't have a cadre of people who were recognized as really competent in the appropriate areas, both in the field and at Headquarters.

    I used to get signals, when I was in NASA, from individuals in some of the centers. I'd sometimes get these messages, and I'd think, I'm just sick of having these guys bug me. And my first tendency was to pick up the phone and tell the center director to turn this guy off. But then I realized, this guy was telling me something important. We'd better take a look at it. So I'd go to the appropriate program office and say, "I've got this information, I think you ought to take a look at it, I'd like to know what you find." You don't have that kind of an alert system in NASA today. To me, that's really a fundamental problem that NASA has.

MAUER: Why?

SEAMANS: I think it's because NASA didn't have the resources to keep that kind of a person interested in staying in the government.

MAUER: So you see some of the problems that occurred with the Space Shuttle program having to do with, NASA wasn't able to keep attracted to itself this same quality of people?

SEAMANS: Yes, I do. I really think that's at the heart of the problem. That was one side of it. The other side was, the pressure to build up the numbers of launches per month. It wasn't that Reagan called Jim Beggs and said, "Look, we're having a convention and I want the launch to take place on a certain date." It was the pressure, somewhat self-imposed, of Jim Beggs and the group trying to prove they could take on the whole launch operation for the United States by putting up all the military satellites in addition to civilian.

MAUER: Why do you think this expertise was lost out? You indicated that even going back into the days of NACA, there was a strong differential, but you indicate that facilities created a work environment. Where did the change come about, and why?

SEAMANS: Well, I think it came about during the Nixon Administration. Nixon wasn't really enamored by the Apollo program, and I don't want to put a political motive on it by saying that Nixon wasn't too enthusiastic about things started by Kennedy, and that may not be the reason. But I know that the research center in Cambridge that was started during the Kennedy Administration was canceled very soon after Nixon took office. And for whatever reason perhaps budgetary. You never can do all you want to do with the national budget even up at the trillion dollar level, choices have to be made. Nixon had his agenda, and it did not include a major space effort.

    He saw the advantage of having a reasonable manned space program. He was out on the Yorktown, no, the Hornet when Neil Armstrong came back, and he could see that there was considerably favorable reaction around the world to that program. He did ask a task force to take a look at what we ought to do next, and Agnew chaired that, and I happened to be on that group. But it didn't take very long going into the Nixon Administration to cut back on the existing Apollo program. He took two flights, lunar flights, out, and started in very sparingly on the follow-on effort, which was the Shuttle. There was really no significant follow-on to Apollo. The Skylab was pretty well under way before he became President. It took a lot to get Nixon to agree to go ahead with the Shuttle, and in that bargaining process, recognizing that the Shuttle was going to take X amount of resources, and Fletcher was having a hard time maintaining a total amount, the tendency was to cut everything else. It's always very hard to justify what appears to be an almost level of effort of research with a vast development of effort chewing up the dollars.

    If NASA doesn't have the dollars to support the research even if they can pay the salaries the people doing that kind of work will say what the hell, I'm being offered a lot more money somewhere else, why should I stick around if I'm not going to have anything interesting to do? You can't keep good people on ice, as it were. Really good competent people are going to gravitate to where the most exciting effort is.

MAUER: Is it to some degree one of the fundamental paradoxes of the large research and development institutions such as NASA-- how do you see the country coming to terms with, you have this large infrastructure, you have to have expertise to be able to make use of that infrastructure, how does a country go about keeping NASA humming without also just throwing money away?

SEAMANS: Well, I don't quite understand what I've been describing as throwing money away.

MAUER: No, no, I wasn't clear enough, excuse me. I don't think it's throwing money away either, but if you only keep capacity just to have it for some unknown future reason, which NASA hasn't been doing, I don't mean to suggest that, but one problem, one worry that budget people are going to have, that when R&D comes to NASA, is, "is this money actually productive that we're giving you?" How do you keep NASA productive but also allow it the space to be able to do what it needs?

SEAMANS: Well, the formal process you go through now is very cumbersome. There isn't any nice simple answer to your very difficult question. I'll say that. I had a little experience-- I will answer your question, just give me a minute to ramble on mentally.

    In the Air Force, when I became the Secretary of the Air Force, we had 1.3 million people in the Air Force, and we were planning to phase down in Southeast Asia, so I could see we were going to reduce the number of people in the Air Force, and by the time I left we were down to 900,000, which is still an awful lot of people, to be sure, of which a certain percentage were civilians, a large percentage military. The only way we can could do that in a sensible fashion was to lop things off. We could not, with such a big organization, just take percentage cuts all over the place. And wherever we lopped something off, it was bound to be difficult.

    I've known two great politicians in my Washington days. One of them was Mel Laird and the other was Jim Webb. Very different people, but very good at understanding the overall context of situations and how to get things done. We had to figure out within all of DOD what bases we were going to give up. We had to figure what future operations would be required obviously. Then in one fell swoop, we'd announce all the bases we were closing, and if we did it on a large enough scale, everybody would tend to get hurt some, so we could push it through, as opposed to closing one base somewhere, where everybody in that geographic area would raise Cain.

    There's a terrible tendency for the political system to give too much ground and they don't accomplish much, and NASA has gotten into that sort of that trap. They've got, just to take two centers, they've got Huntsville and they've got Houston, and there tends to be some overlap between the two. Over time, the centers of NASA tended to adapt to this situation, and realized that if they were just too focussed, then they could be lopped off. So there's a natural tendency to have every center become a little NASA. And now all of a sudden, on a thing like the Space Station, the extreme case of that, development may be initiated in Houston, sent to Huntsville to have something added, and then back to Houston again. The procedures become very complicated.

    But when NASA tried to change in order to simplify, even though it didn't change the manpower very much in those two locations, delegations from each political region came in and raised hell, not with NASA. Hell, they went right to the White House. And at the White House they couldn't understand it, of course, what was really involved, and the tendency there is to say, "Well, this doesn't amount to much anyway, let's just tell NASA to keep doing it the way they were doing it." So there's a tendency to maintain the status quo once you get into this trap, it's very difficult to get out of it.

    So what should NASA do? I would say that somewhere along the line, NASA's got to cut down on their infrastructure, and that's very hard to do. I can't sit here and tell you they ought to close Huntsville or they ought to close Houston. But it's very hard to cut back percentage-wise across the board and still have a viable organization. That's one thing. But no matter how much they have to cut back, and overall they've gone from 35,000 down to is it 25, something like that. They've cut down the numbers of people, but with all the centers still there, you see, NASA still has all the overall carrying charges.

    But what they've got to maintain is a balance between whatever programs they can justify, and I'm talking now about some of the major missions, and a percentage level of effort on advanced work. NASA can't just have the missions and lop off all the advanced work, and still hope to have the expertise there to manage these big projects. They just become a bunch of administrative people without full understanding of what it is they're in charge of.

MAUER: But there's also a potential problem that in the process of lopping off, that you might end up lopping off critical expertise.

SEAMANS: Expertise, yes.

MAUER: And particularly in NASA, where the analogy with the Air Force might break down, and if I'm wrong in this, correct me, is that you have the different missions that the Air Force performs, and then you have quite a few bases that are associated with these different missions, so in closing down a base, you don't eliminate the mission. But in NASA, you have the several major centers, each one of them tending to specialize, so as I understand it in the early 1970s there was a lot of talk going on about perhaps closing down Marshall, and that's one of the reasons why Shuttle was considered so important. How do you do this lopping off in an institution such as NASA?

SEAMANS: Well, first let me say I have not in the back of my mind a plan that NASA should have carried out in '72.

MAUER: Understood.

SEAMANS: And yet also you're absolutely right, that the Air Force with, I don't know, 50 major bases, can lop off five of them a lot more easily than NASA could eliminate one of ten centers, even though percentage-wise it's the same. But let's say, take any one center. Supposing that they felt that they had to eliminate one and it had to be a major one, and let's say they finally zeroed in on Huntsville, and they could maybe justify it in part that the Army there might pick up some of the pieces, and justify it maybe on that basis. They were going to rely more on the DOD for their launch vehicles. But at the same time not wanting to lose all the launch vehicle capability, they would have to take some piece of the effort at Huntsville and let's say move it to Houston. That's not easy to do. But back in '60, we did move a thousand people who were involved in the Mercury program to Houston. It is possible to do it. But what you have to have to do it, whatever the plan is, it has to be done in a grand way. You can't do it piecemeal. Somewhere, out of sight of practically everybody, they'd have to put together a plan, and announce the plan and its rationale, take all the flak and do it. It would have to be so well-thought-out that 80 percent of the people on the Hill would be really for it, so that even though there are 20 percent there from the Alabama area that fight the move tooth and nail all the way, NASA can drive it through.

MAUER: But given the fact that NASA deals with things in terms of specific projects, you have the curve effect. How--

SEAMANS: Well, if you have enough of these, if you phase it right, it comes out more or less constant.

MAUER: Yes, but NASA doesn't really have the control over that.

SEAMANS: To some extent they do.

MAUER: You think they have enough then?

SEAMANS: Well, let's say they have 25 major projects going on, but two or three of them are going to be a lot larger than the rest. They tend to swamp the situation and give the peaks and valleys.

TATAREWICZ: In general, it's usually the tailing? funding curve that for some reason or another won't come down, so the only option you have, the only control you have is control over when you initiate or when you actually move into Phase C/D on your programs, because they tend to stay up there. It doesn't tend to come down as fast as--

SEAMANS: That's true, just like my chart shows. But it isn't just the line--also there's the tendency to keep it up there. You have a marching army, what are you going to do with it? The army personnel ends up taking in each other's laundry after a while rather than accomplishing anything. Well, these are the real issues. And again, I would hope that these oral histories that the Smithsonian is getting into, on this whole side of things, can get at these issues. They're the ones that are really important, for future planners, to get some understanding of how they might deal with them. Jim Webb was very good at it. Maybe to some extent you can say that this whole concept of the Gulf Coast was a lot of malarkey, but it gave him a basis for getting these kinds of decisions accomplished, and I guess you also have to say that it's easier when you're building than it is when you're cutting back.

MAUER: I don't have any further questions at this time.

TATAREWICZ: No, not knowing what Martin has asked in the past and where he stands with this, I'm reluctant to--

SEAMANS: Well, this is just a little background. But let me say where are we with Martin. We're well into the NASA years. We have not discussed the fire and that what followed. And I judge that, as far as I'm concerned personally with my oral history, he's going to spend a little time on how I spent my time in the Air Force and ERTA but not an awful lot. I think he wants to concentrate primarily on the NASA experience.

TATAREWICZ: Well, thank you very much.

SEAMANS: Okay.

Rev. 09/06/96