TAPE 1, SIDE 1

MR. COLLINS: Last time we explored some of the approaches you utilized in managing the Minuteman program. One thing I wanted to look at a little more closely was the way in which you worked with the contractors. We discussed some of the periodic meetings that you held, which I think typically involved contractors. Correct me if I'm wrong, but what you called the "Black Saturday" meetings included contractor representatives?

GENERAL PHILLIPS: No. No.

COLLINS: Just the Air Force and TRW?

PHILLIPS: The Black Saturdays were conducted by the commander of the Air Force Ballistic Missile Division [BMD]. The Black Saturday concept, or that approach to the management at the top level of the Ballistic Missile Division, was started by Ben Schriever when he was commander. At the time I was director of the Minuteman program, starting in the summer of 1959, Schriever had already moved to the Washington area as commander of the then Air Research and Development Command, and the Ballistic Missile Division was commanded by Major General Ozzie Ritland. His deputy commander incidentally was Terry Terhune, who in later years was the deputy out at JPL [Jet Propulsion Laboratory].

But the Black Saturdays were organized and conducted at that level, and their purpose of course was for the commander and his staff to review the major programs. And as I think I mentioned in a previous discussion, my recollection is that generally about a half a day was devoted to the ballistic missile programs, the Atlas, Titan, Minuteman, and Thor, and at various times, especially in later years, there were some other newer programs and advanced development projects that fitted into the structure. For example, for a while there was a very active program for a mid-range MRBM. It was a mid-range ballistic missile, mobile mid-range was its title. That's as an example.

    And the other half day generally was devoted to the space programs, the military space programs. Those meetings were attended by the upper staff of the Ballistic Missile Division, by the top managers of what was then Space Technology Laboratory [STL]--in later years of course became TRW--and also then by the program directors in BMD and their counterpart program managers in STL. And of course there were always certain supporting players there. For example, depending on the subjects to be discussed, I might have two or three of my military officers from the program office there, and other program directors similarly. So the industrial contractors were not present in those meetings.

    But then I think, to go on a little bit to discussing industrial contractors, it was my practice as Minuteman program director to have a management review session. I think the frequency varied at different parts of the program. There was a period when I was conducting them monthly and would have in the project managers from each of the associate contractors. At different times in the program, as I recall, I did this monthly, where each of the industrial program managers were there, and at other times less frequently, like quarterly. So that was one set of meetings in which the industry project people were there, people like for Boeing, T. Wilson, for North American Autonetics, which had the guidance system, Ernie Youngren was the project manager. For the three stages, the first, second, and third stage: Thiokol--with a little effort I could remember the name. He later wound up as the head of Thiokol. And Aerojet had the second stage, and his name was Catrell, and Hercules had the third stage. I don't recall offhand the name of their project manager. And AVCO in the early years had the re-entry vehicle. Those meetings that I just described were the project-level or program-level meetings.

    There was another set of meetings which developed in the latter years of the program, and they were actually started by Tom Gerrity when he became commander of what was by then the Ballistic Systems Division. Those meetings became known as the Minuteman Executives Meetings, and they were not periodic in a predictable sense. You know, they weren't conducted quarterly or monthly or anything like that. They were more nearly when required, but that usually turned out to be about quarterly. And they were actually sponsored and directed really by the commander of the Ballistic Systems Division, who as I said was Tom Gerrity, and they were started as we got into the site activation phase of the program. That was really the production phase. It was the phase of the program where construction work on the sites was heavily under way, and some of the early sites were being finished. The industrial contractors, led by Boeing, who had the contract to manage the site activation, were very much involved in installation of equipment and silos and in the work involving the hundreds of miles of trenches and thousands of miles of wire that had to get laid to wire it all together. So that was the site activation phase, and that was the period when Tom Gerrity felt that getting together periodically but fairly frequently the CEOs of the industrial companies concerned, together with their project managers, together with the key people on the Air Force side, would be useful. And it turned out to be very, very useful.

COLLINS: Was there something more critical about this phase of the project, as you got into the site activation activity, as compared to the development activity?

PHILLIPS: Well, something more critical. That was really where the effort became massive, in a sense. Massive in terms of continuing R&D going on in labs and factories, continuing development, hardware being produced in great quantities. Remember, this was a rapid production program, our purpose being to deploy hundreds of missiles in short order. So the factories were turning out hardware in great quantities. The construction job was a mind-boggling job. You visualize a project where, under one construction contractor, to go into an area around Great Falls, Montana, to build 150 silos, holes in the ground, which were very sophisticated incidentally in terms of their construction, and fifteen launch control centers. It was a massive construction job, and it was a massive job of cable trenching and laying, to put in the network to wire all those silos and launch control centers together in a way that had multiple paths and redundancies. I think of it as a massive job involving a great deal of coordination among all the parties, and a great deal of close management attention and coordination to schedule everything and keep track of it and manage the thousands of people who were involved. And generally it was pretty remote areas, so it was necessary to figure out how people were going to be accommodated to live, if you will. And a lot of the work had to go on during the winter and the severe weather in the northern latitudes of Montana.

    So you asked, was there something critical that caused this to happen. I wouldn't use the word critical. As I said, involving the top executives of those industrial companies together, and together with the Air Force and STL people, turned out to be, I think, tremendously productive. Those meetings were held in different locations. They were held right on site in Montana, for example, at Great Falls. They were hosted if you will by each of the industrial contractors in turn. So this gave the top executives of those companies a personal insight into what was going on, and I can't help but believe that it gave them a personal commitment and a personal involvement which they otherwise would not have had. It gave that whole group the ability to talk to each other and to pick up the phone and talk to each other if there was some reason to do that. So there's just a lot of ways in which it paid off.

COLLINS: Had this device been employed for any of the other ballistic missile programs?

PHILLIPS: Not to my knowledge, and certainly not in the way it was done in Minuteman. That concept was adopted in Apollo. George Mueller, as the associate administrator for manned space flight and my boss, picked up that same concept. He had of course been involved with it, at least peripherally, when he was in STL in earlier years, and in Apollo I think the payoff was tremendous.

COLLINS: Okay. You've outlined the review meetings that you did with the program managers and the meetings with the industrial executives. Were there meetings at a lower level for people who were more intimately involved with the technical issues?

PHILLIPS: Yes, and continuous. I had a relatively small military staff as a program office. I had, for example, a small group of officers under Colonel Dick Hemsley that was identified as the engineering division. Generally the way we worked, I looked to Dick Hemsley to identify military persons. His subordinates were typically, I guess, majors as project officers on the various major pieces of the program. For example, he'd identified a project officer for say the first stage, and on through the program. And those people were continually working the details of that specific contractor and of their counterparts, if you want to think of it that way, in STL, who were responsible for the technical direction of that specific contract. So the people in my program office and the people in STL were--I think of it as just continually--in working-level contact with people in the industrial contractor organization, and that involved travel to the contractor's plant or the contractor's people visiting where we were located in Inglewood, California, and so on.

COLLINS: Were there either Air Force or STL people permanently, or for long periods of time, situated at contractors' plants as part of this action?

PHILLIPS: No. And I'll have to say, there were certain exceptions to that "No." The way the Air Force was organized and operated in those years, each major industrial plant had an Air Force plant rep office. AFPR, they were called, Air Force Plant Representative. And those offices were staffed with Air Force officers and civilians. Local Contract Administration was the broad title of their job.

The one at Boeing, for example, which had existed for years, was fairly large, and their technical know-how was fairly limited at least insofar as the Minuteman was concerned, but their attention to the contract details was in-depth. They were the ones who did the local job of quality control. They had inspectors, and they were the QC people among other things. They were involved in working with our contracting people at Inglewood in negotiating contract changes and working the details of pricing and things of that kind.

    So that was one exception to my "No." Another exception that I wanted to point out when the time is right in this discussion is that I established during the site activation phase--the timing was probably a little bit after the establishment of the Minuteman Executives Meetings--what I called the Minuteman Production Board. Now that was patterned after an approach that had been used years before during the B-47 program, which was a very large program involving many hundreds of airplanes produced in more than one factory, managed out of Wright Field. But I established what we called the Minuteman Production Board, patterned after that B-47 approach, and I appointed Colonel Jim Foster, who was my deputy for procurement and production in the program office there in Inglewood, to head that production board. He physically moved up to Seattle where we decided to locate the production board, and we directed each of the industrial contractors to assign a top-level knowledgeable person to represent their company on that board. And they set up an office or a facility, the space being provided by Boeing, in which each of the major industrial contractors had the data on their production schedules, and they were in direct contact with their factories, and they were knowledgeable in detail of problems that might be occurring in their own factories. The whole purpose of the thing was to coordinate the production and site activation supporting effort, and the multiple detailed schedules that had to feed all of that, and to deal with problems that would occur. As I recall, the principal supporting staffing for that effort was provided by Boeing because they had the contract to pull it all together in an integration sense and to do the site activation management. So that in itself was a pretty massive job, as the work built up at the field sites. Remember, there were, as I recall, a total of five major sites in which Minuteman was being deployed, not all simultaneously, but there was Malmstrom Air Force Base, Great Falls, Montana; Ellsworth Air Force Base, in South Dakota. I'm beginning to--

COLLINS: We can fill those in.

PHILLIPS: Yes. Okay, so there were five bases. As time went on and the program developed, we were working simultaneously in two or three of those major sites, and all of them were being fed by this whole system of contractors. So the Minuteman Production Board, headed by Air Force Colonel Jim Foster, located in Seattle, operated for a few years, and therefore that was the second exception to my "no" about people being located in industrial plants. And just a final comment on that. That was different in later years in Apollo, as you probably already know. The NASA [National Aeronautics and Space Administration] Centers, Marshall in particular, located large numbers of generally technical people in their contractor industrial plants. We did not do that in the case of the Air Force.

COLLINS: But there was frequent contact between technical people on the Air Force STL side and the contractor side.

PHILLIPS: Almost continuous, yes.

COLLINS: What was the nature of your oversight or involvement in that kind of interaction between the STL and the contractors?

PHILLIPS: My personal involvement?

COLLINS: As program manager for the Air Force.

PHILLIPS: Well, it was multifaceted. My own personal approach to managing that project, and Apollo later for that matter, was to spend a lot of time myself at the industrial contractors' plants, at test locations, at the test ranges, both Patrick and Vandenberg, wherever the action was. I spent a lot of time traveling to those locations and then time in turn with the people that were there and the actions that were going on. So in that way, I was involved pretty directly with the project people of the industrial contractor and my own program office and the STL people who were working that particular subject, so that was one way that I was involved with their activities.

    Another way I mentioned in our last meeting. One of the principal devices that STL used in their management of their system engineering and particularly technical direction job were these technical direction [TD] meetings that they held with industrial contractors. Those TD meetings, as they were called, were typically attended by my program office people who were assigned to that particular part of the project. And once in a while, not regularly, I would attend the TD meeting or at least part of it, perhaps because of some critical problem that Bob Bennett or later Bob Burnett had told me about, so I'd sit in on the meeting, or because I would be involved--not always but sometimes--in signing off on the contract direction that resulted from those TD meetings. I've already mentioned my periodic project management meetings with the whole assembled group. Those are a few of the ways at least that I was personally interacting.

COLLINS: Okay. Another way to put it would be, what was the nature of the supervisory relationship or chain-of-command relationship between you and the STL people? Can you kind of describe the nature of that oversight relationship?

PHILLIPS: Yes. It started at the top. My concept and STL's concept was that the STL project manager, who in the beginning was Bob Bennett and a little later was Bob Burnett then for many years, was that that STL program manager was for all practical purposes my technical deputy and the technical director of the program. And as a matter of fact, when we moved our program offices out to Norden Air Force Base, this being in early '62 as I recall, we located our offices in a way that Bob Burnett's office was next door to mine, and we had a door adjoining, and that door was open a good bit of the time. In other words, we were working together in dealing with the top-level problems of the program. So that pattern of the key Air Force officers in the program working with their STL counterparts was similar. In other words, it was more or less patterned after that. I can remember the names of the top STL project people in the various areas. Some of them I mentioned at our previous meeting. They were working directly with the Air Force counterpart in that particular area and in my program office. But when problems were critical, they were meeting with me frequently, generally in the company of either Bennett or Burnett. People like Bob Anderson, when we were dealing with the major problems of getting the engines working reliably--or the rocket motors, I guess more properly called.

COLLINS: This notion of explicit, formalized technical direction was relatively new in the way the Air Force was handling programs. What was the response of the principal contractors to having to work in this fashion and receive this kind of direction on their activities?

PHILLIPS: It was mixed. You're right, it was new. To my knowledge the Air Force Ballistic Missile Program was the first application of this effort, where the Air Force contracted with STL to provide system engineering and specific technical direction, and then inserted the enabling clauses in each of the industrial contracts to require the industrial contractors to respond to that technical direction. So there was a contractual requirement that structured all of this. The reactions, as I said, were mixed. I think after it had been in effect for two or three or four years, in general the attitude in the industrial contractors was that this was a great way to operate. But in the early years, there was a certain amount of skepticism and even resistance, which is, at least to me, I think understandable.

    I can recall, for example, that the key person in the STL program structure who dealt most with Boeing, at the top if you will, was Bill Besserer. Bill Besserer was originally Bob Bennett's and then later Bob Burnett's deputy, so he was the number two person in the STL structure, so therefore he was involved in everything. But he spent a great deal of his time and energy working directly with Boeing because their contract was ultimately the system integration, so that's where a lot of the system attention worked. The relationship of Bill Besserer with the key people in the Boeing structure was, what even today T. Wilson will tell you, was a love-hate relationship. Maybe he's already used those same words with you, I don't know. It was a love-hate relationship. At least the good people at Boeing couldn't help but respect Bill Besserer's knowledge and ability, and in general I think they knew he was right when he told them, by God, do something this way. But there's I guess kind of a natural resistance to being told, especially how to do something. You know, in a very competent company like Boeing, they don't particularly like to be told how to do it. They're willing to be told what to do but not necessarily how.

    Well, all of us were doing a lot of learning in that period. And I can remember several incidents, but one in particular had to do with how cables were constructed and marked and used in one of the engineering model areas or test areas. It was a system test area. And I've forgotten the details, but Bill Besserer was absolutely adamant about how those cables had to be designed and built and marked and used. And that was almost a cause in terms of Boeing's dislike, if you will, of being told exactly how to do all that. But in the end Bill Besserer turned out I think to be right in insisting how that testing was to be done. As I've said, I've forgotten the details, but I remember it was a major case.

COLLINS: In general, in a situation like that, where there was a difference of opinion about how a technical option ought to be handled, how would the difference of opinion be resolved?

PHILLIPS: Well, there were a few, and if the difference of opinion was big enough, it came to me. One of the first cases I got involved in I mentioned to you in our last discussion, but just to identify it again, was a strong difference of opinion between STL and Boeing over how many tests we needed to conduct in that tethered silo launch series at Edwards. That one was a big enough issue that it did come to me, and really in the final analysis it was presented to me by Bob Bennett on the one side and his people, and T. Wilson and his people on the other side, and I made the decision as to what we were going to do. That's I guess probably the first one that occurred. I know there were others that were big enough and perhaps contentious enough that they came to me. Usually there was no delay in getting matters that needed that kind of decision up to me if it needed that kind of decision.

COLLINS: We've talked about several management mechanisms that were employed in working with the contractors. I guess I'm curious to know, in your view, which one of these falls under the rubric of technical direction. You had a specific set of technical direction meetings that were employed, but you also had these other types of review meetings. Would you categorize any of those as part of the technical direction effort?

PHILLIPS: Not in the sense that the technical direction was defined in the contracts. In the sense that in a top-level program review meeting, one which I conducted a significant technical issue was reviewed and even decided and directed, that was certainly technical direction. I think what I'm trying to differentiate is the technical direction that was specified in contract to be done by STL, and they did that, generally at their level, if you will, through the means of technical direction meetings, and visits to contractors, and their devices. In a program review meeting, as I mentioned, where a major technical matter might be decided and directed, the STL people were present, and their voice was certainly heard in the process of that action. So I guess in that sense, clearly it was technical direction. I guess I used to think of it more as just program direction as opposed to specifically technical direction.

Since you bring up this subject of technical direction, I think you have to recognize that the ultimate responsibility for the conduct of the program was the Air Force's, and I had been assigned the job of managing that as the program director. I think probably in one of our early discussions, I mentioned that in my own career case, I had felt strongly that a mix of Air Force operational field experience in operating commands, together with R&D or program acquisition management experience, was valuable.

    And a case in point here. My assignment immediately preceding the Minuteman program direction job was as director of logistics and materiel for SAC Seventh Air Division in England. That had been three years of field Air Force operating experience. Well, when I came into the Minuteman program, I found that the design of the system and the missile under a concept which made sense, called the wooden missile, was headed toward building that missile and delivering it to SAC in a way in which the re-entry vehicle and the guidance system could not be removed in the field. I may have mentioned this at an earlier meeting.

COLLINS: Let me pause here while I turn the tape over.

TAPE 1, SIDE 2

PHILLIPS: We're talking generally I guess about technical direction. I was discussing the design of the Minuteman missile as I found it when I first came into the program in the summer of 1959. The design at that point, as I said, contemplated building the missile in a factory, assembling it in a factory, in a way in which the guidance system and the re-entry vehicle could not be removed from the missile by SAC, if you will, the field operators. Although a lot of the details of the design still had to be worked out, it was clearly intended even in that wooden missile concept that there be some way in which the nuclear components of the re-entry vehicle could be removed and replaced or worked in the field. But the principle I'm getting at is the concept that the missile was not to be disassembled in any sense or any fashion in the field.

    Largely based on my previous three years of experience and responsibility in SAC, I knew that it would never work over a period of years for the re-entry vehicle and its nuclear weapon to be inaccessible, if you will, to SAC. Because the typical pattern in SAC for nuclear weapons of course involves the storage points and very specially trained munitions units that are trained especially in how to handle that, and there are very sophisticated rules that have grown up over the years about how that will be done. So as I say, I knew it wouldn't work to build the thing in a way that it wouldn't fit into that general pattern. And related to that, I knew also or believed that the guidance and control system, which was near the top of the missile, a section as I recall approximately three-feet long that included the guidance and control equipment and key electronics, that that section should be removable in the field for maintenance work.

    And so I directed that there be field maintenance joints, if you will, built into the missile in a way that the re-entry vehicle could be removed completely, and the guidance system could be removed completely. Clearly that was a technical direction. It was also a major program direction. I didn't feel it was necessary to do a whole lot of coordinating on that particular decision. It was not resisted in any major sense by STL or anybody else. I think of it I guess as a violation of or a deviation from the original concept of a wooden missile.

COLLINS: I have not heard that expression before, wooden missile. What does that refer to?

PHILLIPS: Well, it was a fairly generally used term in some of the early planning of Minuteman, and it was intended to connotate a missile that didn't need maintenance. You remember that one of the real objectives of Minuteman was to build and deploy a large force that required relatively few people, and that of course required the remote control, unattended sites, very, very long periods in which no remove-and-replace was required. And the original concept under this wooden missile concept was that a missile would be built in the factory, that it would be immediately transported to and installed in a silo, and that if any part of it ever required, if something failed in it, the whole thing would be removed and taken back to the factory. That was the original wooden missile concept. Of course wooden missile is perhaps in a way a misnomer, because it was quite a sophisticated machine with a lot of electronics that were running continuously.

COLLINS: But the basic design decision, to make the re-entry vehicle and the guidance inaccessible, stem from this original notion of a wooden missile as a key part of the system.

PHILLIPS: Yes. Right.

COLLINS: Did your decision then result in a kind of modification of that concept? Did this require more people to be at the site to be involved with maintenance issues, or did this have any implications for that?

PHILLIPS: Well, it had implications, of course, because the way that we proceeded then with the design was--this is going back quite a while in memory now. I'm trying to remember exactly how we installed the first missiles or the early missiles in the sites. I think I'm correct in my memory that the missile was assembled with all of the components except the re-entry vehicle, was assembled and checked out in the factory at Ogden, which was operated by Boeing. It was an Air Force owned, contractor operated plant located on government property adjacent to Hill Air Force Base at Ogden, Utah. That was the assembly plant for the Minuteman. And the missiles were assembled there and then transported by rail to Montana for the first wing and other sites for later ones.

COLLINS: This was the complete missile?

PHILLIPS: Yes.

COLLINS: Including warhead and all?

PHILLIPS: No, the re-entry vehicle was not on it.

COLLINS: Okay.

PHILLIPS: I'm quite sure in my memory that the checkout in the factory included the guidance cam, the guidance section. So that part of the missile then went into the silo, and that dictated the design of the transporter erector where the missile would be transported out to the site and then erected and lowered into the hole. Of course, that then dictated that we have a special vehicle to bring out the re-entry vehicle with its nuclear warhead and be able to lower it onto the missile and install it and connect it and test it. It also dictated that we have a vehicle and people who were trained to do this, to remove the guidance-and-control cam and to transport it back to a facility that we had to design and build, where it could be tested, and certain remove-and-replace possibilities were built into it, I think. But at least the ability to replace the whole guidance and control system and to test the whole thing in a facility was built into it all.

COLLINS: So this would be done on some kind of periodic basis?

PHILLIPS: No, not periodical, only in the case of failure. See, the design requirement--I don't remember the exact numbers--on the guidance and control equipment was for it to be able to operate continuously, which it had to do on strategic alert in the silo, for over a year. There is an exact number but I just don't remember it. It was specified in hours. And so the intent here was to design and build this thing in a way that it would operate continuously without attention for a very long period of time, over a year. And the concept even today I think, but certainly in the early years, was that when the guidance system failed, and typically they were the unit that failed first, that guidance cam was removed. In other words, they had to send out the RV vehicle to get the re-entry vehicle and its warhead off and take care of it, and then the next vehicle came out and took off the guidance cam and took it back, and a replacement one that had been tested and checked out in the facilities on the base was sent out and installed.

COLLINS: I assume that most of the missiles at any given base were installed at close to the same time, so that in likelihood all the things would be approaching a probable failure of their systems at approximately the same time. How did you grapple with that kind of problem, if that's an accurate characterization?

PHILLIPS: Well, Malmstrom, which originally had 150 missiles-- I've forgotten the time interval from when the first one was installed to the last, but it had to have been at least a year. So over a period of a year 150 missiles were installed and brought up. It could have been a little longer than a year for all I remember. How did we deal with that from a management standpoint? The first effort was to continue to drive hard to get the reliability up and the lifetime of the system. And that involved a continual push, particularly with Autonetics, the contractor, to identify failure items. As the operational experience started to build up, we got data, so over a long period of time there was a continual feedback, if you will, to continue to improve the reliability and endurance of the system. So that was one of the main thrusts.

    The second main thrust had to do with how many spare systems we provided. That was an effort which, one end of the title of provisioning, spares provisioning, and that was a pattern of activities that went on in the early stages of the program and involved people from SAC as the operating command, from what today would be the logistics commander responsible for spares and maintenance. Back in those days it was the Air Materiel Command. And our R&D people and our contractor people would work the question of how many spares should be provided on what schedule. That was a pretty massive effort, incidentally, and went under the title of provisioning.

Early in the program, a decision was made for the Air Force to establish its own depot capability to overhaul and maintain those guidance and control systems. That facility was located in Ohio, and it was in a facility, part of which was underground. I keep getting off on interesting sidelights. Right after World War II, there was a big effort directed out of Wright Field to provide very heavy hydraulic presses for aircraft manufacture. The purpose being, with these very large, tremendous presses, to be able to stamp out major parts of airplanes, bulkhead sections and skin sections and so on. There was more than one location in the country where these tremendous big presses were going to be installed, and one of them was a place in Ohio, and I keep trying to remember the name of the little town. For that purpose, a fairly tremendous base was established underground, tremendous mass of concrete and steel to provide the geodetic stability required for this monstrous machine with its motions and so on. Our guidance experts said that that's the world's best location for the facility in which we will test our stable platforms. You know that thing's not going to move. Where the heck was that? It was in Ohio. Anyhow, that's where we built a depot that still operates. So any guidance and control system that failed in the field and was not locally repairable, and I don't think that there was very much that they did with those in the field, would go back to Newark, Ohio, to that depot. And that's still in operation, I'm sure.

COLLINS: So these things were refurbished and then put back in.

PHILLIPS: Oh yes.

COLLINS: This seems like a pretty different orientation from the wooden missile concept. I assume that under the wooden missile concept, when a guidance system failed, you had to replace the whole rocket.

PHILLIPS: Yes. And that's what I saw, or believed early on to be a fallacy, and therefore directed that the guidance system be removable. The rest of the missile for all practical purposes then followed that wooden missile concept and generally was durable for years, endured, in other words didn't fail.

And generally the replacement of those up through the three stages was a time-based thing, based on the growing experience in the life of those solid propellant motors. For which, incidentally, we established a major facility under the Air Force depot at Hill Air Force Base at Ogden, Utah, for life testing of solid propellants, and that facility probably still operates even today. From the first of our production program, which would have started in '60 or '61, somewhere in that period, into this propellant life test facility there would go a sample from each motor produced, and there would be a random selection of motors that would be put into that facility, and they were then kept under controlled conditions and periodically measured and evaluated, and periodically a sample would be taken out and fired. The purpose of that was to determine what the life of the solid rocket motor was. Our original design requirement, if I remember it right, was something like three years. Ultimately the life of those rocket motors proved to be much longer than that, and I wouldn't be surprised that some of the motors in the system today have been around for fifteen or so years. But that's the kind of in-depth program I think is required in order to establish what's safe and what the reliability factors are and what the replacement cycles ought to be.

COLLINS: Just to be clear now, for the guidance thing, it wasn't a question of just periodic replacement based on some sense of when things might fail, but was there some kind of active checkout of the guidance systems to determine a failure?

PHILLIPS: Yes, the latter. It was not a time-phased replacement of the guidance system. It was when it failed. The guidance and control system in the Minuteman missile was running continuously twenty-four hours a day, because the requirement of the system was to be on strategic alert at all times and to be capable of being launched within thirty seconds of the signal or the command to launch. So it was on alert, thirty seconds to go, twenty-four hours a day, for years. So the design of the guidance and control system was such that it could do that but also it was continually self-testing, and in the event of a failure internally, that was identified and signalled out, if you will, to the launch control center that there was a malfunction or a failure, at which time the launch control officer would tell maintenance to go deal with it.

COLLINS: Did you get involved at all with questions of how potential failures of different parts of the Minuteman system affected strategic questions? How did one relate the reliability aspects to what you could effectively plan strategically and think about?

PHILLIPS: Well, the answer is, yes. I was involved in a number of meetings with the SAC people in particular. The SAC people of course were the ones concerned with the ultimate operational employment. So they had to have their own planning factors as to what the reliability was in the silos and in flight, in order that they could do their own operational planning. So they had to come up with a number say for a given squadron of fifty missiles: how many would they expect to be or could they plan on being capable of firing if the signal were given?

    You know, in current years I suspect that number is very close to 100 percent. In other words, that system has been perfected to the point where the reliability of missiles that are supposed to be on alert is almost 100 percent. In earlier years, I don't remember what the number was, but it was 90 or 80. In the early years, there were more failures than in later years. It was always high but in the early years, certainly it was somewhat below 100. The planning factors for reliability in flight came largely from flight-test data, the flight testing in the latter stages of flight testing being done primarily out of Vandenberg. So it was primarily Vandenberg launch data that went into building up the statistics of in-flight reliability which SAC used. I was involved in structuring the various engineering-related activities that would provide these kinds of data. I can recall a number of meetings with SAC people to work those questions, and to understand what it is that they required and to be able to accommodate it within our efforts.

COLLINS: I'm kind of groping here but I'm interested in what the relationship is between people with the operational planning interests, and what they feel they need to achieve, and somebody like yourself who is responsible for the technology, how that interaction works and what's the nature of the give-and-take there.

PHILLIPS: Well, it was different in the ballistic missile program than in a typical Air Force program, first of all. I may have touched on some of this earlier. At the time the ballistic missile program was decided and directed and established in mid-1954, it was a top down decision and direction. I recall we did discuss this I guess last time. In that very early period, Curt LeMay as commander-in-chief of SAC was very opposed to the ballistic missiles. I think he saw it as draining off money that would have gone into his bomber programs. As I say, LeMay was very opposed to the ballistic missile program early on, and I think I mentioned that in the initial establishment of the Western Development Division, the first name of the Ballistic Missile Division, the WDD was assigned the responsibility for creating the initial operational capability in the ballistic missiles, and that had something to do with the origins of Vandenberg, for example. And of course the first of the ICBMs was the Atlas. Well, by the time the Minuteman program went under contract, which was the fall of 1958, SAC had come around and was a party to the ballistic missile program. And by then I guess, by '58, sites were being built for the early Atlases, for example, on SAC bases and SAC people were being trained.

    So I set the stage this way in order to indicate that the ballistic missile program was different, in terms of the typical relationship between an operator and a developer. And so the Minuteman, by the time I took it over a year after the contracts had been let, this being in the summer of 1959, the interactions with SAC were fairly minimal at that point. I know that the majority of the planning and of the system engineering and the design as it existed at that point had been done with essentially no input or interaction from SAC. The top-level requirements had largely been developed in interaction with the Ballistic Missile Committee, the top of the Air Force, the Secretary, Chief-level. The top-level performance requirements of fifty-five hundred nautical miles of range, the specific accuracy--which I don't know whether it's still classified or not but it was a specific number--and certain of those top parameters were clearly directed but not really worked out in the usual sense with SAC at all. Well, I guess maybe my assignment to take over the Minuteman brought what I think of as the first significant involvement of SAC thinking into the Minuteman, with direction like I told you about, to make the RV and the guidance sections removable.

COLLINS: Just a quick question here. Was this more a matter of your experience, having worked for SAC and understood their frame of mind and approach to things? Or was this part of a general change within the relationship between SAC and the ballistic missile division?

PHILLIPS: The former. It was just directly a result of my own experience in SAC. I ought to respond to questions. I talk too much. What's next?

COLLINS: I interrupted you as you were starting to talk more about the relationship--you were saying that with your involvement the interaction between SAC and BMD increased.

PHILLIPS: Right. Well, Okay. I don't remember the exact timing, but really in my time at BMD, SAC established an office there. It was called SAC-MIC, and I don't know why that name but I guess it was SAC-Missiles. So they established a liaison office, and they had a few officers assigned out there to be liaison with BMD, and fairly early in my tenure, as I say, this office was being established. I've forgotten exactly how it occurred, but I wound up with what I wanted an identifiable SAC group that were a part of the Minuteman program office. And they were of course part of this SAC-MIC organization, which was not large. It was initially probably only three or so people that were assigned to me in the Minuteman program. So they were part of SAC-MIC and then in turn through them reported to SAC in Omaha and did a lot of the coordination between our program and SAC.

    Another thing that I don't remember exactly the origin of, but I visited SAC fairly frequently, probably at least not less often than every six months. Maybe more often. I just don't remember exactly. To brief the SAC staff and LeMay on the program and its status. Those meetings, which had to have started not later than some time in 1960, were kind of a top- level kind of meeting in which the top SAC people were being briefed and informed. But by then, the requirements on SAC for their planning were beginning to build up because, you know, they had to do their own base planning and their own people planning,and so the involvement got to be fairly heavy, probably starting in about 1960. But by then SAC, as I said, had clearly embraced the ballistic missile program, and it was a part of their structure, and there was not resistance to it. And their influence then began to be felt more and more on details of how things were designed or planned, like the facilities in which the RV would be maintained. They had a heavy involvement in specifying what was to be constructed and how it was to be equipped and so on.

COLLINS: When were the RVs mounted onto the rocket? I'm unclear on how that part of it worked. When the missile was shipped from Ogden, it included everything but the RV. What happened after that point, once it got to the site?

PHILLIPS: There was and is some facility at each of those operating bases for the missile, as it would be delivered from the factory, to at least reside until it was required. There was, to my memory, no checkout and no maintenance planned to be done on the missile, as it would arrive or be involved with the base. So there was some kind of holding or transfer facility. Typically they came in on a rail, and they had to be loaded onto the transporter erector to be taken out to their sites. I know there had to be some kind of holding facility, surge facility if you will, to hold them until they were required. In the site activation phase, my organization was responsible for completing, accepting for the government in a contractual sense, and then turning over to SAC the sites with missiles a flight at a time, something like that. So if you visualize that what was being created and turned over to SAC then was a set of ten silos with missiles in it, including guidance systems all working and checked out and so on.

    SAC then was responsible for getting them on strategic alert, and they had their own requirements to follow in order to do that. But from a technical standpoint, a design standpoint, with an exception I can discuss if you want, they were designed so that you had to have at least two launch control centers involved to be able to launch them. After the initial twenty, then you can add them ten at a time. Well, so, on SAC's own schedule, they would accept them into their structure, and whenever their own operational requirements dictated, they would assemble and test an RV and warhead in the munitions facility that had been created for them, transport it out on the transporter that we had provided, install it on the missile, and do whatever final check was required, and close up the silo, and it was on alert.

COLLINS: So SAC took care of loading on the re-entry vehicle?

PHILLIPS: Yes. The only handling of that warhead was done by SAC. Well, that's enough said, I guess.

COLLINS: Okay, I'm at the end of a side of tape here.

TAPE 2, SIDE 1

COLLINS: As you were beginning to activate the sites, or perhaps even before, was there a concern about the vulnerability of the missile sites, of the silos? I know that became certainly more of a concern later on, but I wondered how early that kind of issue came about?

PHILLIPS: Well, almost from Day One in the initial design, and continuously through the program. That concern had led to the initial design, certain features of the design, initially to include the fence, the area surveillance system. In other words, there was a system installed that would detect motion or presence anywhere inside that fence, and many other features of the design were dictated by that very early attention to making the sites sufficiently invulnerable. I've forgotten the exact timing. Our initial operational date was October of 1962, so it was probably in 1960 some time when the interaction with the people at SAC and the interaction with the people in the Pentagon raised the level of concern to a fairly high level, and we were concerned about a number of things. I remember that we let a contract with Bell Laboratories to do an in-depth assessment of the vulnerability of our whole launch control system, to accidental or inadvertent or saboteur or unauthorized party being able to launch the missiles. So that whole contract with Bell Labs was oriented to assessing the accidental, inadvertent, unauthorized launch problem. And they did a comprehensive job, and that was part of our actions that were devoted to concern being expressed at appropriate different levels over this problem, and also by our own awareness that we were going to have to pass a test at the end, which was for the President of the United States to sign off on this system being okay. And in the end President [Dwight D.] Eisenhower was involved in doing that.

    The concerns that we had, and which were then worked in detail by Bell Labs, ranged all the way from the vulnerability of our total cable plant--remember, we had hundreds, maybe even thousands of miles of buried cable hooking all this together--and concern that somewhere out in some isolated field in Montana, somebody could dig down and get at our cable and hook onto it in a way that they could do all kinds of bad things, including launch missiles. Our initial design had contemplated some of this, or I guess maybe even most all of it. The cable itself, which had a large number of conductors in it and was fairly large, was pressurized with gas so that any penetration of the outer sheath would presumably release the pressure and give an indication that something had occurred to get into the cable.

    Our initial design of the launch control system had also been based on designing it in a way that it required pretty sophisticated digital codes to be transmitted from a launch control center, or a pair of launch control centers, to a missile, before the signal would be accepted as a valid launch signal, which incidentally was my introduction to the National Security Agency [NSA]. We were proceeding full blast on our own and had designed our launch control system and its digital code generators and all that stuff, on our own. I don't remember exactly how it occurred, but somewhere along the way the National Security Agency became aware of what we were doing and made it known that they had the mandate in the federal government to provide codes and coding devices, and we were in effect out of line by doing what we were doing.

    So I mounted a meeting at Fort Meade. That's where I first met Lou Tordella, who was at that time the deputy director of NSA. I guess Gordon Blake was director at that time, and in previous years he'd been my boss, so I knew him. Tordella lives out here in Maryland, getting on in years, but in later years when I was director of NSA, he was my deputy. I got to know him very well then. But to make a long story short, we worked out an arrangement with NSA, starting with that meeting where I and several of our people went to Fort Meade to work this problem, where NSA had a proper role in our launch control system coding and decoding and the associated designs, so that they could play their proper role in that.

    Well, the Bell Labs work essentially validated that we were okay, insofar as our cable plant and all of the security involved there, but they pointed out that if someone were able to get inside a missile site, inside the silo, that there were certain places in which they could, with relatively simple connections and equipment, cause a missile to be launched. So we took what they found or what they pointed out to us, and mounted an expedited design effort to make all of those places that they had identified invulnerable or protected against that. We in the end built certain of the cable runs and certain junction boxes in which wires could be accessed in a way that they were potted with a compound that included Carborundum, so that even with cutting tools, a person could not penetrate to be able to get at the wires.

    So we fixed all those places that they had identified as weaknesses, but in the process of doing all that, we realized that although our missile silo had been designed to be hard against an atomic blast, that a sophisticated penetrator equipped with what you could put on a truck could get inside that site. And our criterion was to have the site resist penetration for thirty minutes, that being more than sufficient time for the security forces to react to the signals that would tell them that the site had been violated.

    So I mounted another special project, that we gave a title called Project Button Up, and the purpose of that project was to identify any weaknesses where the site could be penetrated. For that purpose, we got Boeing and other contractors--I've forgotten, besides STL and some consultants. Those consultants included what were probably some of the most knowledgeable safecrackers in the world, without necessarily clean records in society, plus people from Army arsenals that were experts in shape charges and things of that kind. And the rules we set up were, they could have anything they could load on a 6x6 truck and whatever number of people they wanted, and our problem was to make our site so that they could not penetrate it within thirty minutes.

COLLINS: By penetrate, you mean get into a silo?

PHILLIPS: Get inside the silo. That sophisticated safe cracker penetrating crew was amazing. You know, they could get up to our combination lock that we'd put on our personnel access hatch, that was something like maybe, 2 1/2 or 3 feet in diameter, where people could go down the ladder to get in, but they could work that combination in a matter of literally seconds, or violate that combination lock system and be down through the personnel access hatch. They also had a backhoe on their truck, and they could get in there and start digging. And then the shaped charge experts were putting shaped charges on the wall of the silo at a thin point near the top where they could blast a hole and get in.

    So we built some special test sections at Vandenberg to work all this, as part of this Project Button Up, and initially they were inside our site within milliseconds or a few minutes at the most through the personnel access hatch system. To fix that, we designed quickly what I called a plug, which amounted to a cylinder that essentially filled the cylinder that was the access tunnel, vertical, made out of concrete and steel, mounted on a jackscrew that was provided by Western Gear Company, I remember. We hired Diebold Safe Company, and out of all that effort, we came up with a whole new system of locks at the top, and a second hatch that had a whole new system of combination locks, which gave you access then to this plug, and there was a special set of combinations you had to work in order to make that plug drive itself down. Remember, it was just a big mass of concrete and steel. So it had to be lowered down to the bottom so that a person could then get into the equipment bay at the top of the silo. There were some sections near the top of the silo where shaped charges could let them in, and we beefed up that upper section with concrete and steel. And there were probably other things. But that whole Project Button Up was done in a matter of weeks, and the appropriate design changes were then directed into the system.

    In parallel with all that, and as kind of a final step, that really more or less came out of the final Pentagon reviews of the Bell Labs work, was a decision to add one more layer of launch control safety, which we called the Launch Enable System. I still remember Bob Burnett and I getting on an airplane, going out to Seattle, and meeting with Ollie Bealieu, who was kind of the chief electronic designer at that point in Seattle, and the three of us sitting in Ollie's office there and working out the design requirements for this Launch Enable System or launch safety system, which amounted to adding one more layer on top of the launch control system. For that, there had to be again a special set of digital signals sent that would in effect unlock the system so that the launch control system could then send its signals through, so there was one more layer on top. Boeing then proceeded rapidly to design and test that system and get the design changes then directed into the system. That whole set of efforts that I've just described was done in a matter of several weeks or at most a very few months. It was accommodated into our efforts in a way that it got done within our schedule, without stretching it out, if you will.

COLLINS: I'm unclear about the timing of the Bell Laboratories study. This was before October of '62?

PHILLIPS: Oh, yes. As I think I indicated, this whole effort had to have been some time probably in 1960. It was prompted by reviews and, you know, expressed concerns. I don't know exactly where they all came from. Some of them were internal to my own organization. Some from Omaha, some from the Pentagon. That's what caused us to mount this special review effort, and then that in turn prompted the design change effort. And in the end the results of all of this were a factor in ultimately the White House signing off on the system as being safe.

COLLINS: You've described concerns about intrusion to the base proper by individuals. Were there evolving concerns about what the requirements were for survivability from a nuclear hit?

PHILLIPS: Well, they were continuous also. The very first conception of Minuteman had contemplated the requirement to be invulnerable to an established level to a nuclear blast. By now I think the numbers are unclassified, but the initial requirement was that our launch sites should stand one hundred PSI overpressure, and our launch control centers three hundred PSI overpressure, and that in turn then dictated a lot of the details of the construction. That concern over survivability was continuous from the early concept all the way up to today, and over the years led to a lot of testing.

    When I went back to what was then Samso, after Apollo, this being in late '69, I found that there was a major test effort going on under the direction of the Weapons Lab in New Mexico, involving people from the Atomic Energy Commission or somewhere where they had built the headworks of Minuteman silos of that time, which had evolved in those years, and were subjecting them to realistic overpressures with massive amounts of conventional explosives. Gee, they built a monster out there in the way of a tremendous rack and pile of TNT. God, it was massive, and it was designed in a way that it would create this overpressure, and it was by then a lot more than one hundred PSI. So the concern was continuous. And really in recent years, that's been the driver, of course, in MX basing, "Peacekeeper" as it's called.

COLLINS: Did the requirements for meeting a specific threshold of PSI change at all while you were in the program, or did it stay at those levels?

PHILLIPS: No, it stayed at those levels during my time. But almost from the beginning--I've forgotten just when it was established. I guess it was probably toward the latter part of the period when I was Minuteman program director, which would have been in '62 or '63. By that time there had been a committee established, I guess under the Air Force Scientific Advisory Board, to probe the details of vulnerability and hardness and survivability. And I remember the Latter brothers from RAND were key on that, Al Latter and his brother, Dick Latter. Edward Teller was involved on and off and a number of others. They were a committee of very, very well-educated and knowledgeable specialists in nuclear effects and so on. I've forgotten exactly when they invented it, but they came up with EMP somewhere in that period, electromagnetic pulses. As I said, I've forgotten exactly when EMP became a known factor. But they were all over us all the time on nuclear vulnerability. Blast waves, radiation, EMPs, you name it.

COLLINS: By "they" you mean?

PHILLIPS: "They" being this scientific advisory committee. It went into business--it had to have been '62 or '63, somewhere in that period, and it probably exists continuously even right up to today.

COLLINS: I think we've covered Minuteman pretty well. Is there anything else that you can think of we ought to put down here at this point? Is there anything you think we've left out or didn't cover as thoroughly as we might have?

PHILLIPS: Well, there are a lot of interesting stories about the evolution of that design, from the initial design up through the years. I don't know that we need to go into them now. I guess suffice to say that by the '62, '63 period, the involvement of SAC was growing and considerable for obvious reasons, so they were having more and more influence on the evolving design. I say evolving because model changes over the years were the way of life. Starting probably in 1962, we were laying down the design requirements and design for what became Minuteman II. Remember, Minuteman I--it wasn't originally called that but that's what it later became--had its initial design requirements, and I remember talking about the phase that led us to in effect breaking Minuteman I down into Wing I and the rest of Minuteman I for the reasons of the shortfall in range of the initial delivered missiles.

    I guess there are two or three things I ought to say that were effective in that period. From early on we had a mobile version of Minuteman. We may have talked a little bit about this before. And by probably some time in 1960, we were moving along quite well in developing a mobile version to be put on railroad cars. I know I remember talking a little bit about this, to say that the mix that had been decided on was to have a third of the Minuteman force be mobile, and two-thirds in the hardened and dispersed sites. President [John F.] Kennedy was elected in 1960 and took office early in '61. He appointed [Robert] McNamara as his Secretary of Defense, who appointed Harold Brown as his DDR and E. I've forgotten now whether we talked about this, but fairly early in McNamara's time and fairly early in '61, he visited BMD to review the ballistic missile programs. We did talk about this, didn't we?

COLLINS: Not in this detail, no.

PHILLIPS: He had a number of people with him--I think we did talk about it--including Harold Brown and Johnny Foster, who was at Livermore Labs at that time. Jim Webb, who had just been appointed NASA administrator had been invited by McNamara to come along. Incidentally, George Brown, who later became chairman of the Joint Chiefs, was McNamara's military assistant as a two-star general. Well, in a whole day set of reviews, BMD reviewed all their programs: Atlas, Titan, Minuteman. The most time, as I recall, was spent on Minuteman, with me briefing the whole program. I remember, after I finished my briefings, which included both the hardened and dispersed and the mobile systems, at a break, talking with McNamara as we were looking at a model somewhere, where he was in effect complimentary of the program, how it was going and so on, including the mobile system. The next week, after he and his team got back to Washington, some directives came out canceling several programs. He, on that same trip, had visited Wright Field and reviewed a lot of their programs. And among the programs that were canceled by that following week was mobile Minuteman. I've forgotten whether it was exactly the same day or time, but Dyna-Soar, Skybolt were also, both of those were canceled in that same set of orders. Dyna-Soar was the Air Force's initial effort for what would look like a Shuttle, to be launched on a Titan III. Remember that?

COLLINS: Yes. Did that fall under the BMD auspices?

PHILLIPS: Oh, yes, that was a BMD program, the early space program. Skybolt was an air-launched ballistic missile, to be launched out of a B-52 as an intercontinental ballistic missile. Well, the thing that really hurt me about the cancellation of the mobile Minuteman was that it was announced for the wrong reason, namely, that the program was in technical difficulty, which it by God was not. Of course, I realized at least in later years that a President and a Secretary of Defense have to make their decisions about their defense strategy and the weapons required to carry it out, and all within a budget structure, and it's to me understandable when decisions are made to either do or don't do a particular program, including say the mobile Minuteman. But I've always been resentful of the fact that it was announced to the public as being for the wrong reason. I think the reason to me was fairly clear, namely, if you visualize in the sixties, especially in the later sixties, nuclear warheads on Minuteman missiles riding around the country's railroad tracks, what kinds of problems that would have led to. So I could clearly see a lot of political and social reasons why that program should be canceled. Well, enough said.

COLLINS: I guess that raises a question about the decision-making procedure for these kinds of decisions. It sounds like McNamara came, and he listened to you and then went off and made a judgment based on that presentation. When you were making that presentation, were you aware what the stakes were?

PHILLIPS: No. Well, no, not, I think, in the sense that you asked that question, because there had not been a lot of debate and uncertainty over the correctness of the way we were going. That was new. And it was a new administration.

COLLINS: Right.

PHILLIPS: So it was a surprise when the mobile Minuteman was canceled. I find it ironic and interesting that we're now in the latter stages of proceeding to put Peacekeepers on railroad cars. Incidentally, they've gone back and resurrected some of the things we had already created at Hill Air Force Base in the way of railroad capability within the Air Force. They've gone back and resurrected some of that in the Peacekeeper how many years later? '61 to '89.

COLLINS: Right. So after this briefing, you had no further opportunity to make a case for the mobile Minuteman.

PHILLIPS: No. No. I think another thing I should mention is that with the Kennedy-McNamara administration, the involvement of the Office of the Secretary of Defense [OSD] in details of the Minuteman program grew rapidly. We had people from OSD out there in our Minuteman program office, to me in recollection it seems almost continuously, working details. And one of the themes they picked up on that drove a lot of effort had to do with penetration, with the ability of re-entry vehicles to penetrate enemy defenses. By that time, some evidence I guess had developed that in the USSR there were efforts to develop antiballistic missile systems. So we started then in '61 to do a lot of experimental work and development work and start to feed requirements into Minuteman to develop penetration aids, or to enhance the ability of a re-entry vehicle to penetrate enemy defenses. And it involved decoys and ultimately led to multiple re-entry vehicles, and that in turn led to multiple independently guided re-entry vehicles.

    Also, that continual involvement of the OSD staff people under McNamara with us, I think led to a change in the strategic approach of the country, from what was I guess initially called-- I'd have to check to be exactly sure of the titles--massive retaliation to flexible response. And so we were the guinea pig-- not the guinea pig--we were the action vehicle in that flexible response conversion. Our initial design under the massive retaliation concept, the initial design of our launch control system, believe it or not, launched those missiles in blocks of fifty. In other words, they were all preprogrammed, pretargeted, and when the whistle blew or whatever, those missiles were launched in blocks of fifty, fifty at a time. And that, as I think back and all, was a pretty narrow view for design but that was the way it was initially designed. And I think, and probably properly so, that sort of thing is what led McNamara and the administration at that time then to change from pure massive retaliation to a more controlled and flexible response.

    What did that mean to us? It meant requirements to redesign our launch control system, so that we could launch those missiles one at a time. It led ultimately, and this was a feature in Minuteman II, to multiple targeting, so that you could put more than one target into a missile, and you could change the target under certain conditions without having to go out and remove the guidance cam and all that kind of stuff. So there was quite an evolution then that started, and it impacted us in terms of detailed design, some of it fairly early. I've kind of forgotten, but we got that individual missile launch in pretty quickly. I can't remember now whether we got it in our initial ones or not, but we got it in awfully fast.

COLLINS: It would have been a fairly major redesign, I would think.

PHILLIPS: It was. It was. One other thought, and maybe this is the final one. By late in '63, which is when I was winding up my time as Minuteman program director, Secretary McNamara had reserved to himself decisions on changes involving more than ten million dollars, which up to that time I had been delegating to people working for me in the Minuteman program office. Minuteman could not have been done if the management approach from OSD that came into being early in '61 had been in being in the late fifties. Minuteman could not have been created.

COLLINS: In the sense, there were simply too many layers of decision making involved?

PHILLIPS: Not too many layers. There was too much micromanagement from OSD. It wasn't the layering as much as, as I said, decisions that were being made by captains and majors in '58, '59, '60, '61, '62, were being second-guessed or made either by people in the OSD staff or by McNamara himself.

COLLINS: I'm at the end of a side of tape here.

TAPE 2, SIDE 2

COLLINS: Your last comment about the increased involvement of the Office of the Secretary of Defense in looking at the details and making decisions in the program, I wonder what that means for someone who's a program manager for a project like this? Is it taking too much prerogative and authority out of your hands? What is your sense of the implications of that kind of approach to managing a project such as Minuteman?

PHILLIPS: Well, I've already said that in my view, the Minuteman program could not have been accomplished successfully under the environment and in the way that things were being done by 1963. Why? Too much authority had been withdrawn from the program director and reserved to the Office of the Secretary of Defense. That's an oversimplification, I know. I've thought a lot about this subject. It is necessary for the top-level people in an administration to insure that military programs of all kinds are compatible with the strategy and objectives that that administration has set for the country. I mentioned as one specific example the transition from, I'll call it pure massive retaliation, to a more controlled flexible response in strategic matters, and of course over time that evolved to controlled escalation, and their thinking about how nuclear war might develop, and so on. So my point is that it is necessary for a President and his immediate staff to make some decisions about national policy and national direction and strategic programs and so on. I think the system has to accommodate those decisions being directed where they are carried out. In this case we're talking about a department of defense.

    One way to do it, of course, is for the performance and design requirements to be worked out and directed on a doing organization, and then for an assignment of responsibility and authority and the establishment of a sufficient check-and-balance review kind of procedure to occur. That's essentially how I think the ballistic missile program, from 1954 up into the '63 period, was accomplished, and the doing organization, which was BMD and its program offices, was clearly accountable, and there were sufficient review processes and checks and balances to keep good track. But in general, we were told what to do and left to do it and expected to get it done, and if we had to deviate or if new direction had to be provided, there were expedited means to do that.

    Well, by '63, with the approach of the McNamara administration--I said a while ago, no layering, that isn't necessarily true--but that approach out of OSD clearly required the air staff to be very much involved. All this having started in '61, the air staff was very much involved, and so by then we were in effect going through the layers. I think we talked the other day about the Ballistic Missile Committee, and how the AFSC headquarters and the air staff were being bypassed. Well, as the McNamara administration came into being and as their approach to life became effective, the air staff felt they had to be more involved. So starting in '61 and clearly by late '62, we were going through the regular Air Force structure. The first stop in Washington was over at Andrews for AFSC and the second stop was what was originally called the Weapons Board, the Air Staff Board, and the next step was the Air Force Council, and possibly beyond that was a meeting with the Chief and the Secretary, and then beyond that would be OSD, so you know the layers had started to build. And in that kind of structure, it's maybe an over- exaggeration, but you just can't get things done.

COLLINS: I guess we can continue to explore this. I'm interested in what that means in an organizational sense, not being able to get things done. Is it just that you're expending too much of your energy in grappling with the bureaucracy or reviewing various decisions with numerous people?

PHILLIPS: Yes. You're prevented from deciding and acting on too many issues, and are required to present those matters and the decisions to higher headquarters, and back in the McNamara period that involved OSD to make certain decisions, as I said, changes involving more than ten million dollars. You have to go up through that whole layer with that whole story, and for anything of any significance, that required the program director and a briefing and all that time. So it's a matter of withholding too much in the way of authority, withholding and constraining.

COLLINS: Was this something that you think--this is purely speculation on your part, I assume--was this a strategy on McNamara's part to indeed slow things down, to give him an opportunity to review more of what was going on?

PHILLIPS: I don't know. I don't know. It was maybe a natural part of seeing what I'll characterize as being a good program, well run, but with some design features that had to be changed, like the launch control system and like re-entry vehicle penetration aids and some of these other things. I finally have remembered the name of McNamara's guy who practically lived with us. It was Marvin Stearn. He was the head of strategic matters in I guess the DDR and E staff, probably under Harold Brown. Marvin Stearn and one or two of his helpers were the ones that practically lived with us, working these problems I'm talking about. You know, I hadn't thought of Marvin Stearn or seen him in years, and about six or eight months ago, I ran into him on an airplane traveling to LA.

COLLINS: This is a question that leads toward our discussion of NASA. Did you have an appreciation for what McNamara's overall interests in space were and the various potential programmatic activities that the Air Force was already engaged in and in the CIA? You know, what was he willing to support and hold onto and really wanted to make the prerogative of the Department of Defense?

PHILLIPS: I really don't know, Martin. Ben Schriever would have that insight. I don't. I really don't.

COLLINS: Okay. I think the time is getting close. Do you have to just walk across the street?

PHILLIPS: Yes, I've got another half hour, if we want to use that much.

COLLINS: Why don't we go a few more minutes. Actually I'd like to take you down to the gallery and show you where we've got this Minuteman guidance section in one of our galleries, and just take a look at it. I'd like to get some of your comments on it.

PHILLIPS: Good.

COLLINS: Let's look then at how you came to leave your role as Minuteman program manager and go over to NASA. How did that come about?

PHILLIPS: Most of my answers turn out to be a bit long. You want the long answer?

COLLINS: Sure.

PHILLIPS: NASA, which had been formed from the old NACA and the [Wernher] von Braun part of Redstone Arsenal and JPL, had a tremendous technical and scientific capability, but relatively little to none in the way of experience in managing programs and especially large programs. The management experience was brought into NASA from industry and later from the Department of Defense. You know, the initial efforts in these would have been under Jim Webb, bringing in people like Bob Seamans from RCA, a little later people like Joe Shea from then STL, a little later George Mueller, who came from STL, and Bob Young, who came from Aerojet. So you know, there was an effort to bring in people from industry who had at least industrial management experience in programs, and George Mueller was brought in by Webb in--

COLLINS: 1963.

PHILLIPS: Yes, I was going to say early '63. And George himself laid the groundwork for a program management, program office, program director concept of managing the Apollo, and the Gemini and other programs for that matter that were under his jurisdiction. My own interpretation of the events was that George Mueller, interacting with Seamans and ultimately with Jim Webb, believed that some of the large program management experience in the Department of Defense would be beneficial in managing the Apollo program. So my own interpretation of events is that Jim Webb went to McNamara and in effect asked for help in the way of experienced program manager assignment. I never have heard who all was considered, if there were others. I don't know. I guess I never thought to ask and I really don't know. But McNamara I think agreed to Webb that he would provide somebody.

    Well, the selection process wound up with me being selected, and this was done in some combination of Webb, Mueller, Zuckert, who was Air Force Secretary, and Curt LeMay, who was Air Force Chief of Staff. My first word of the whole thing was a telephone call from George Mueller, who I guess was telling me of the Apollo need, and I guess trying to prepare me for what was coming, which was a decision for me to be transferred. I was initially reluctant. This telephone call from George Mueller had to have been somewhere in October, November of '63. As I say, I was initially reluctant because by then I could see myself becoming the commander of the Air Force Ballistic Missile Division, or what was by then called BSD, Ballistic Systems Division, with you know, an opportunity to go on from there. So I could see a good career path for myself from where I was, staying in the Air Force. I also had some concern about an assignment outside the Air Force and what that might do to sidetrack a career. I must admit, I wasn't initially perhaps thinking very much about how exciting it might be to be in charge of a program to get people on the moon. So that's kind of a long answer to how I got assigned. So specifically, I was ordered, I think with an effective date of the 1st of January of '64, to report on detached service to NASA, and my initial assignment was as deputy program director. Initially George Mueller retained his name as Apollo program director in the acting, in the order chart, and I was deputy. That persisted for not very long. I've forgotten just when but some time during '64 I guess George and Seamans and Webb were satisfied, and so I was appointed as the full Apollo program director.

COLLINS: What were your contacts with George Mueller previously, when you were in the Minuteman program, and he was at STL?

PHILLIPS: Almost none. I hardly knew George Mueller. My perception was that George Mueller was somewhere in the STL staff, and I never interacted with him. I had met him, but I really didn't know him.

COLLINS: You mentioned you had this one meeting early on in the Kennedy Administration that involved Webb and McNamara, during this review of the Minuteman program. Did you have any subsequent contact with Webb after that time, before you came to NASA?

PHILLIPS: No. No, but I'm satisfied that Webb's sitting with McNamara in that daylong review of BMD, that Webb formed his own impressions of people, so he surely had something in mind when it became either apparent, or he and Mueller decided, they were going to go to the Department of Defense to try to get somebody with experience.

COLLINS: Yes. During the period of the early sixties, when it was clear that NASA was going to grow as an organization, what was the attitude within the Ballistic Missile Division about Air Force people going over there and assisting NASA at getting started?

PHILLIPS: I don't remember that that was a particular subject of discussion or concern, one way or the other. Up to the formation of NASA, with the Space Act of '58, there was a lot of infighting going on about roles and missions. That infighting was going on between the Army, Navy, and Air Force, among them, and also in, if you will, debate and opposition against the formation of a new organization. Ben Schriever is the guy who can tell you an awful lot about that period, or Rube [Ruben] Mettler. From my level, or from my knowledge, I guess I'm aware that Eisenhower is the one who made the decision that there would be a civilian organization established to work space for peaceful purposes and so on, which was behind the Space Act of '58. So a lot of fighting went on up to then. I was not myself aware of much thinking or debate about Air Force or military support of this new NASA organization.

    But your question prompts me to pick up with my assignment to NASA. You know, I came into NASA, I think of it as late '63; my actual official reporting date I think was the 1st of January, '64. There were a few people in NASA already who had come from the Department of Defense. A number of the astronauts of course had come from Air Force, Navy, Marines. There were a very small number of Army people who had sort of come along more or less I guess with von Braun, people like Rocco Petrone, Lee James. I think they had as Army officers sort of come in along with the von Braun organization. The Navy I guess had loaned NASA a very small number of people in earlier years who came in as Navy officers and then retired and stayed with NASA, people like Bob Freitag, whom you've probably run into. I think Bob came in as a Navy captain and then retired and stayed with NASA until recently. One of the people I inherited in the Apollo program office, Jack Holcom, I think had come as a Navy captain and retired and stayed on. So there were a very few already there, for various reasons.

    It only took me I guess a few days, it wasn't very long, to kind of size up things, to reach the conclusion for myself that I needed some help in the way of experienced people. So I went back over and met with Curt LeMay and in turn with Gene Zuckert, and told them that I needed help, and explained the situation enough to justify that. They approved it in principle, said, "We'll help as we can," and they gave appropriate direction I guess to the air staff, and as a result I very rapidly worked out a deal for the assignment, relatively quickly, immediate assignment of fifty-five Air Force officers. What I felt was needed were people with big program experience who could go into some of the key positions, not only in Washington but also in the centers.

    And I had worked that a little bit at least with people like Shea and Gilruth and Petrone, and the people at Huntsville, von Braun and his people. I'm sure that my assignment was met with a lack of enthusiasm by a lot of people within NASA when it occurred. There was less than hearty enthusiasm about my bringing in more, but a willingness. It varied. Von Braun wanted help. Gilruth, Shea and company didn't. Petrone had an open mind. It was that sort of thing.

    But I wound up getting a number of people that had worthwhile experience, and I rejected a bunch, incidentally. Ed O'Connor, who was by then I think he was a brigadier general, went in then as von Braun's deputy to head the program side of Huntsville. Remember, I think we talked about von Braun organizing within what they called industrial operations, which was the program offices, and their technical operations, which was their laboratories. Ed O'Connor headed the industrial operations, all the way through Apollo. So the program managers working for him--Arthur Rudolph on the Saturn V, Lee James on the Saturn 1B, Lee Beaulieu on the engines--they all reported to Ed. We got Dave Jones, who was by then I guess also a brigadier general, and he went in as a deputy of sorts to George Mueller. I've forgotten the details, but I felt that this request justified also having somebody up there to help George. With a little more thought I could name a lot more, but I got a chief of program control, a colonel named Roy Seckem with experience, so I put him in charge of program control. A lieutenant colonel named Jack Collippe and whom I put in charge of working out our configuration management and basic system engineering requirements, and let's see, who did we wind up sending to Houston? George Abbey was an Air Force captain initially. You know George, don't you?

COLLINS: No, I don't.

PHILLIPS: George retired out of the Air Force several years ago after coming in through this program of mine. He was George Low's righthandman during George's time as Apollo spacecraft program manager. In later years, he wound up in charge of the flight crew division. I think he was over here as a deputy to Dick Truly until Dick moved over to the administrator's office. I think he's over here now.

    Well, my point there is that I did get the Air Force to assign on detached service fifty-five Air Force officers, ranging from one-star generals down through captains. They were assigned in Washington and in all three centers in positions where their program experience could contribute. My own view is that they contributed a lot to the conduct of the Apollo program. I would be interested in the views of others, looking back. From that early beginning, other Air Force officers were assigned at various times in the following years. And at the peak of the Apollo program, we had on the order of four hundred Air Force officers assigned to NASA.

COLLINS: Okay. I'm thinking maybe we should break off here today.

Rev. 09/06/96