Interview with Harold "Doc" Irving Ewen

Description

Harold "Doc" Irving Ewen, 1922-2015. Interviewed 12 August 1979 in Weston, Massachusetts, length of interview: 135 minutes.

Creator

Papers of Woodruff T. Sullivan III

Rights

NRAO/AUI/NSF

Type

Oral History

Interviewer

Sullivan, Woodruff T., III

Interviewee

Ewen, Harold I. "Doc"

Location

Weston, MA

Original Format of Digital Item

Audio cassette tape

Duration

135 minutes

Interview Date

1979-08-12

Interview Topics

His thesis with Purcell on search for 21 cm HI line (1948-51) and work as "engineer" for Harvard 21cm group (1952-57); US radio astronomy versus others, industry and military versus academia; "style" of science and engineering.

Notes

The interview listed below was either transcribed as part of Sullivan's research for his book, Cosmic Noise: A History o Early Radio Astronomy (Cambridge University Press, 2009) or was transcribed in the NRAO Archives by Sierra Smith in 2012-2013. The transcription may have been read and edited for clarity by Sullivan, and may have also been read and edited by the interviewee. Any notes added in the reading/editing process by Sullivan, the interviewee, or others who read the transcript have been included in brackets. If the interview was transcribed for Sullivan, the original typescript of the interview is available in the NRAO Archives. Sullivan's notes about each interview are available on the individual interviewee's Web page. During processing, full names of institutions and people were added in brackets and if especially long the interview was split into parts reflecting the sides of the original audio cassette tapes. We are grateful for the 2011 Herbert C. Pollock Award from Dudley Observatory which funded digitization of the original cassette tapes, and for a 2012 grant from American Institute of Physics, Center for the History of Physics, which funded the work of posting these interviews to the Web.

Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event.

Series

Working Files Series

Unit

Individuals Unit

Transcription

Partially transcribed for Sullivan by an employee of Ewen-Knight. Transcript completed by Sierra Smith in 2013.

Sullivan

August 12, 1979, I’m talking with Harold "Doc" Ewen at his home in western Massachusetts on the early years of 21 cm line but before we get into that can you please tell me what your educational background was, and how it was that you first came into contact with astronomy and what came to be known as radio astronomy?

Ewen

I graduated from Tech High in ‘39 in Springfield [Massachusetts], and went on to American International College as a freshman after deciding not to go to Springfield College and become a muscle man. [Ewen: Refused 4 Year Full Tuition, National Hi-Y Scholarship] Then I went to my sophomore year at Amherst. Actually I was on my way up to UMass and thought I was at UMass when I stopped at Amherst and went in and tried their swimming pool. And it turned out at the time I was swimming with the son of one of the professor in astronomy at Amherst. And he suggested that it was a short walk from the swimming pool to the Dean’s office and we became close friends immediately. And the Dean then asked me that typical question, "Why do you want to come to Amherst?" And I told him that I wanted to teach at Amherst and I thought that anyone who was a teacher at Amherst should be also a graduate of Amherst. He said he hadn’t heard that one before and based on that answer he offered me a full tuition scholarship at Amherst. So I entered as a sophomore and that went on through the junior year until the war broke out. At that time I elected to accelerate my program in the summer of '42 to graduate then in January of '43. Also at that time pressures were put on the faculty at Amherst to call into... essentially, analytical duty at MIT in the Rad Laboratory. And they lost one of its professors, John Hall, who later became the senior observer at Kitt Peak or director at Kitt Peak. John Hall at that time was at Amherst and when he left, then began a search to find a substitute teacher in the field of astronomy. They located me as a senior and I had at that time, just the prior a couple of months, had decided also to major in astronomy. So I was offered the opportunity at the end of my senior year to teach astronomy as a member of the faculty, and also then finish up in January.

Sullivan

So you fulfilled your wish earlier than you might have thought about teaching at Amherst?

Ewen

As a matter of fact, it was done very cleverly by President King, who called me into his office one day. In the spring of ’42, I was just finishing the junior year and reminded me of that answer that I’d given the Dean. And he said, "I’ve got the note here and I’ve also got the recommendations for you to be placed on the faculty as a senior. So here’s your chance." So that’s how I joined the faculty. One of the problems, of course, was at the commencement, would I walk in with the faculty or my classmates. And I guess the record that was sat was the youngest member of the faculty in the 122 years of the College at the time. But that was my history at Amherst and I stayed on at Amherst for one year as an Assistant Director of the Navy V5 training program. That was important for the occasion primarily of working with a director who was head of the Astronomy Department, Warren Green, also the head of the Astronomy Department. Warren Green and I became very close friends during that year while we were handling the V5 Program but it was historical also from the standpoint that the Navy sent their baseball cadets to Amherst and so I had their students for about six to eight months: Ted Williams, and Johnny Pesky, and Johnny Sain. And I got to know them pretty well and everything I’d give Ted a tough time on an exam, he’d take me down to the ball cage and explain to me that he knew how to pitch as well. If I would just stand there at the plate, he would throw a few over and see if I could drive them up into the net. So I’ve been a good buddy of Ted’s some while and it was handy later because many of my scientific friends are also friends of the baseball thing. And I recall on one occasion where Oppenheimer dropped into the lab at Harvard, we all came to attention, shuddering graduate students, and said, "I’ve got a question for you guys. Where’s Fenway Park?" So that was the sort of thing that was very good there. Now from there, we go from Amherst then into World War II with the Navy training in radar electronics, Princeton, MIT. Then off to air training in Florida and Chincoteague, Virginia, and ultimately to join a Patrol Bombing Squadron 112, which was at that time located in England. It had moved up from Africa. And so during that period when I was with the Patrol Bombing Squadron, my assignment as a Radar Officer flying with the squardron, was to fly with the eighteen aircraft, patrolling the English Channel. So we spent a fair amount of time off the Netherlands, and France, and Germany looking for submarines at fairly low altitude in a Liberator type aircraft. It was coincidently about the same time that [Hendrik C.] van de Hulst was giving his presentation in the Netherlands on the 21 cm line and also the foundation of all of the lines that would be forthcoming in the historical development of radio astronomy.

Sullivan

So, you were by the end of the war then very familiar with radar both technically and operationally?

Ewen

Yes. I came back from the Navy, from my last post in Washington, to Harvard to enter the graduate school for a doctorate in physics, which I had selection primarily because the interested in the atomic bomb as a benefactor of my being able to return from anywhere, since I was presumably to go from Whidbey Island to Attu. That had been explain to all naval air types, that the weather conditions on Attu were such that you took off and had a choice of landing in Russia or China after you bombed northern Japan. It was a one flight per shot and you had trained for two months for your one flight. It was in that period that we heard about Hiroshima and Nagasaki and it changed my thinking about my future education, other than finding out what these fellows had uncorked back at the labs in Cambridge. So I then did apply, as a matter of fact, with the Office of Research and Inventions, which became ONR. I had selected a position there as a personnel officer, in the office of personnel, just relocating officers here and there, rather than work as a so-called electronics expert for a patent attorney. And when the time came for filling out applications for the future, Professor Warren Green, again from Amherst who was quite disgusted with my interest in business and law and things like that that I chatted with him about, sent me an application for an NRC fellowship in physics because I told him, "I was now interested in physics but not astronomy." And so he sent me the application, I filled it out, sent it in, and that upset ONR because their physics department, astronomy department- research, there was only one NRC fellow and he happened to be in administration that year. So with my NRC fellowship under my arm, I went off to Harvard to study physics.

Sullivan

This was ’46?

Ewen

This was in ’46, the fall of ’46. And the initial time at Harvard was spent getting the necessary courses out the way. My problem primarily was that I was short in the study of physics. I had not had a good background in classical physics, because my work at Amherst had been predominantly mathematics and then briefly astronomy and only then the application of mathematics to astronomy. So fellows like Newton and so forth were not that well known to me. And it became an interesting game at Harvard during those early years, ‘46-‘48, that I found it easier to understand relativistic theory and quantum mechanics than understand how I could make N larger and agree with those fellows that liked to bounce apples off heads. So I became known fairly shortly as the "quantum baby". If you take a fellow right out of the womb and put him into graduate school and teach him relativity, he will say, "What else is there?" And it was just so natural and so, but I would occasionally stumble given a problem in classical physics. I had very little difficult with the quantum mechanism and standard thermo and so on.

During that period because of the Navy policy, I was retained as a naval officer on inactive duty. But I would put on my "blue suit" and run down to the Squantum Naval Air Base or to the Fargo building to meet with the Harvard Group, as it was called, of reserve officers every couple of weeks. Our CO at the time was Donald Menzel who during the war was a four striper [Navy Captain] and after the war, he was elevated to Rear Admiral. So problem then was how we could do something productive that would relate to Menzel's interest in the sun, our interest in whatever we were studying at the graduate schools in the local area, Harvard and MIT. And I came up with the thought of building a solar radio telescope since that seemed to be in Menzel's area of interest and it happened to be something that I didn't think would be at all that difficult from what I had heard concerning solar signals.

Sullivan

Where had you heard about this?

Ewen

I learned about the detection through Menzel when he gave a talk at the Reserve Officer meetings. And by then, in preparation for the Menzel seemed mildly interested in this and said, "What do you really know about the field of radio astronomy? It’s come a fairly long way. Are you familiar with Grote Reber and all?" And I didn’t know anything. So he said, "Why don’t you spent your two weeks of active duty at the [Sterling Labs?] with Grote Reber and I’ll make the arrangements." So, Don Menzel made the arrangements for me to go down and spend a couple of weeks with Reber, which were very productive in understanding of how far away to stop a Model- I can’t recall, a Model type car I had at the time. It wasn’t the Model A I had at Amherst. But Reber was quite concerned with ignition noise and things like that. It became quite a buzz word in the whole field. So you usually walked into Reber’s "camp" from quite a distance. So I spent a couple of weeks with Grote Reber and became somewhat fascinated with what he was doing.

Sullivan

What were your impressions of him at that time? Can you remember?

Ewen

My impression was that he seemed to be- either considered me as though a grad student. I think it did help a little in his opinion of me that I ultimately made arrangements for the Navy to wear civilian clothes rather than the uniform. The uniform he found very appalling and he would not talk to me and he just said, "That’s stupid. Take that off." So I finally [?] upon the Navy and they said, "Yes. If you are working up there in the woods with Grote Reber, that is fine." And then he would talk to me a little but I had the impression that he was racing against time. He seemed to be overwhelmed with an important task to be completed, which he alone understood. And he was not in a position to describe in detail what the plan would be even for the two week period other than, "Follow me. Pick up that screwdriver. Hold this meter. Do that." And I felt like a real "gopher running around, trying to catch on to what was going on in his mind and why he seemed so desperately anxious to consume every minute. He was always there bright and early in the morning. He worked late. Forget about meals. He was just an around-the-clock, not a [gadgeteer?] at all, just fellow trying to get his job done. It was a very motivating experience. I think became friends to the degree, at the time of parting, of sort of going to a summer camp experience, sort of a mild respect not knowing each other well enough, but understanding the space that surrounded both of us. I recall his saying goodbye and I had to come up and see him in my uniform because I was leaving to go to good old Boston and he said: "You try to get out of that as soon as you can." At least he talked to me for the first time in uniform.

Sullivan

What was he doing scientifically at that time?

Ewen

Primarily, at that time as I recall, his main interest was solar. Though I talked him about a year ago, he said that at that time was trying to get a better feeling for improvement in receiver technology. He had a number of Würzberg antennas there, and there were certain obvious things that he felt could be done, and he hadn't discussed them in detail with the Bureau [the National Bureau of Standards, NBS]. As a matter of fact it was about that time, I think it was within a week or two after I left to give you a feeling for what was on his mind, that he disappeared and they tried to send him the famous check that came back [Sullivan: because Reber took off so suddenly and left no forwarding address]. Then finally, some months later he showed up at Maui.

Sullivan

I hadn’t heard the famous check story.

Ewen

He just didn’t show up and no one knew where he was and the checks came back.

Sullivan

His paychecks?

Ewen

Paychecks. They were just coming about to the Bureau. Where is he? So it got to be, "Has any one seen Grote Reber?" And then about six, seven months later, somebody saw him with a bucket of bamboo on top of Maui. And then he disappeared from Maui and showed up in Tasmania. So he has these moments. And I happened then to see him at the Bureau just prior- an interruption on his way to Maui. But I must say, I don't feel that he has really changed all that much. Just in the last year when I saw him, he was the same exciting fellow, eager to learn and so intense.

Sullivan

That's why I was interested in your reaction. At that time- well hold it now - we're talking about '48-'49?

Ewen

We're talking about '47- maybe, about in there. I was in the Reserve. I left the Reserve in '48, I believe.

Sullivan

Are you getting mixed up with another visit with him as far as him taking off? I think he stayed at NBS for a few more years, until '51 or '52.

Ewen

He went before I left Harvard, I know that.

Sullivan

Yes. That's true.

Ewen

I really wasn't in that close touch with him. I guess it's probably a little difficult when you're a grad student, because I wasn't in that close touch with radio astronomy.

Sullivan

In any case, what I was going to ask you about was his interest in detecting the 21 centimeter line. In fact, he has pictures of his equipment that he was putting together in the late ‘40s.

Ewen

His comment on that, as a matter of fact last year, was that he took a look at it, and he was going to try for it, but he was pretty sure that it was going to be so difficult from a radio standpoint, that it probably wouldn't be worth the engineering effort and I agree with him.

Sullivan

But did you talk to him at that time at all about...

Ewen

About 21, no, because it hadn't entered my life at that time. His interests, as I recall, were primarily solar, and galactic mapping of course, but we didn't talk about any line work. And then as you will recall, to put it in context, during that period and right up to '49, my interest in radio astronomy was equivalent, or motivated, or whatever, by the requirement that I report to the Navy one night every two weeks. So, one night every two weeks I would get interested in radio astronomy with my "Rear Admiral", and we would talk about the sun and other good things out there. Then I'd disappear for two weeks and go back and do something important like nuclear physics.

Sullivan

How long did this go on?

Ewen

Oh, this went on until '49. In '48 I decided to take a position at the Nuclear Cyclotron Lab and went down to Washington and chatted with the administrators at NRC [the National Research Council] because you are not supposed to work while you're a graduate student. So we negotiated and I elected to give up the fellowship [Sullivan: and his Naval Reserve position also?] and proceeded to work at the Nuclear Lab.

Sullivan

What's the official name of that lab?

Ewen

It was called the Nuclear Cyclotron Laboratory.

Sullivan

So, this was a 50 inch cyclotron?

Ewen

A 95 inch.

Sullivan

A 95 inch cyclotron that was being built at Harvard at that time.

Ewen

And we completed it in '49 and wrote it up in 1950. Then I actually went to work experimentally- carrying out experiments. My assignment was to obtain an external beam. Another contribution I meas was details on frequency modulation of the cyclotron itself, moving the capacitive or variable portion of the circuit outside of the magnetic field. During that period I had completed my master’s work, I had completed my course work, and I’d become quite fascinated with the field of nuclear physics, and this great gadget, the cyclotron. And it was time then to seriously think about a doctorate thesis. Of all the teachers that I had encountered, the one I found most fascinating was Ed [Edward Mills] Purcell and that was through a course in electron ballistics because I had just a slight feeling for electronics from the Navy. A crash program of 90 days at MIT is good, but not an EE [electrical engineering] degree. And then through the experience with Purcell of taking a ride on an electron through a magnetic field and discovering that you are dealing with just about any type of electronic device in a very simple way. It was such a pleasure, very much like his current book where you take a "relativistic ride". So, I found a sort of companionship or sincere interest in his way of looking at things. So I then selected him as my thesis advisor without a thesis and we agreed to sit down and think about thesis subjects. My suggestion package was that it would be helpful to me if I could take the various areas of application, focus them in one direction. There was the radar work I’d done, astronomy, math, because Ed asked, "Well, what are you proficient in?" And I told him that I had just a smattering of knowledge of a number of "mountains" and I felt pretty comfortable in the "valleys", and I notice that there are a lot of fellows on one "mountain" that don’t talk to others and if I could just play in that game. I said in just working with a group in the Naval Reserve, Menzel’s group, that by the time I knew a little bit more about quantum mechanics and related that to the astronomical observations, I could see the dichotomy of the radio astronomer from the astronomy end of it, and the electronics engineer, and that they weren’t really talking on the same wavelength.

I thought that perhaps there was a place in the valley to bring all this together. And I thought a very interesting one would be to do a bang-up job on the [22 GHz] water vapor line, because I taught meteorology while I was at Amherst. And I had taken while I was at Harvard- Harvard had a requirement in the Physics Department, that you become, in effect, a master of four subjects. That’s one of their Physics Department requirements. You could satisfy two by courses, you could satisfy a third by taking an oral examination at a master's level and the fourth would be your doctorate field. And at first with some amazement, I didn’t realize that there were four separate fields of physics that you could get involved in but apparently Harvard had that approach to it. And so I picked math which was accepted and quantum physics and then elected to take meteorology as a third subject and it required the permission of my thesis advisor. I talked to Ed about it and he said, "That’s a pseudo-science, not a science - We don't even teach it here." I said, "They teach it down at MIT, and I've been taking courses down there and I haven't decided yet, but I'm so close to completing all courses in meteorology at MIT that I may elect to take a doctorate in meteorology. I think it is sort of a science, if we can separate the various parts of meteorology, the understanding of circulations, and things like that - it's certainly physics of some sort." That was before aerodynamics was a popular thing. Harvard at the time had the [?] Observatory and Professor Brooks and that was Harvard’s total part of the program, as a contribution to meteorology.

Sullivan

So you were actually thinking about being a doctor in meteorology?

Ewen

I really didn't know what I wanted to do.

Sullivan

But that a distinct possibility?

Ewen

Right because I taught meteorology at Amherst, but that was when I was one chapter ahead of the students, and I thought it would be really nice to know a little more about it. And so I did then during that period take the courses at MIT, during ’47, ‘8, and so on for a doctorate. So then my oral was in the area of meteorology.

Sullivan

What were you thinking of doing with the water vapor line?

Ewen

My interest there was really to get a better feeling for the variability in water vapor, since what [Robert H.] Dicke had done was to demonstrate a technique, but he did not have the opportunity to milk the science that might be there.

Sullivan

Then he didn't get much information on the atmosphere?

Ewen

That's true. And so with the meteorology, and with this electronic technique, it seemed to marry a few disciplines and it also marrying Dicke to Purcell which I found interesting. When I first looked up Dicke's paper and found Purcell's Ok [Sullivan: authorizing] signature on it, that helped a little in focusing my attention on the water line at 22 [GHz]. So I suggested that to Purcell and he didn’t seem to be too enthusiastic about the idea. He said, "It would just be another point on a curve that had been done." It could be done. He wasn't refusing me the opportunity, but that it might take a considerable amount of time to measure the line structure in adequate detail to represent a useful thesis.A couple of weeks later when we were still thrashing around from a program, I was over at his house and he said, "I know you are interested in that Dicke thing but it occurs to me..."

Sullivan

Why don’t you just finish this story.

Ewen

He said, "A fellow, I think he’s Dutch, I’m not sure, but I saw a paper the other day or an abstract where a fellow has apparently come up with a prediction that you can measure the hydrogen hyperfine line."

Sullivan

In interstellar space?

Ewen

In interstellar space and he said, "You ought to look into that," and I said, "He’s from where?" He said, "Well, he’s a Dutchman I think, and I saw it somewhere." He said, "I read so many doggone things that come across my desk but I thought of you when I saw it because that’s got to be a lot better than doing a water line in the atmosphere." And I said, "Well, what’s the probability that it can be detected and why doesn’t he do it, why doesn’t this Dutchman do it?" And he said, "Well, I don’t know. Those are things that you need to look into," and he said, "But I’ll tell you, if you can get a gas line in interstellar space, anything like that, not the atmosphere. But if you can get that, you’d get your face in Life Magazine and no one would ever forget you." And he said, "That would be a real great point on the curve and if you don’t, you’ve only wasted a couple years of your life." He said, "You're the kind of guy who likes to do different kinds of things. Why don't you do this thing? Give it a whirl." And that was his pitch. "What have you got to lose? Just a couple of years."

Sullivan

What year would this be?

Ewen

I would say the fall of '49.

[Lunch Break]

Sullivan

So, continuing after lunch now. So having heard this, apparently you had some interest and then what did you do, how did you proceed with the problem?

Ewen

I did a typical literature search, located a copy of van de Hulst's paper, and arranged to have it transcribe by Purcell's wife, Beth, who was studying Japanese and Russian at the time, and was capable of transcribe just about anything.

Sullivan

Translate you mean.

Ewen

Translate. Translate. We had that in hand and, then as part of the so-called literature search, I began going around to the various club meetings of astronomers, the American Astronomical Society [AAS], etc, and also discovered that Commission V of URSI [International Union of Radio Science] was far more popular from a radio astronomer's standpoint because in the astronomy field, the optical astronomers considered radio astronomers essentially as second-class citizens. And Commission V of URSI was the first receptive international or even national audience. It wasn’t that it was international, that wasn’t at all important. There just was no place for radio astronomers to go to publish and discuss things.

Sullivan

So, you were going to URSI meetings while you were doing your thesis work before '51?

Ewen

No, this was in early '49. This was shortly after finishing off the oral [Sullivan: exam] in meteorology, and the course studies, and so forth. Looking into this essentially in the direction of Purcell to find out more about the Dutch paper and also about the field and just get a feel for it.

Sullivan

What I'm getting at is that you did get into radio astronomy before 1951. It wasn't until '51 that you began finding out about American astronomers?

Ewen

Oh no. Not at all. I had done quite a bit of reading and had become familiar with Reber's work.

Sullivan

And NRL’s work?

Ewen

Absolutely.

Sullivan

Now, did you get any useful advice, during these couple of years before '51, about radiometers or astronomy, as far as that goes?

Ewen

None at all. As a matter of fact, I was pretty much alone as far as the radio astronomy part of it was concerned, because there was no activity at Harvard in that area, nor at MIT to my knowledge. So I would depart from this community of Cambridge scholars, so called, and then go off to these meetings and come back to report, "There are people out there working in this field." And so, nor did I make any close allies or friends in the field at these meetings because they were rather short with me. Though I did get one occasion to chat with John Hagen, who is a Boston-type, who was at NRL at the time. And I recall leaning over at some meeting, it was down at Yale, Commission V URSI, and I tapped him on the shoulder and said, "I’m Doc Ewen and I’m from Harvard and I was wondering where, when you are speaking later this afternoon, could you talk about this 21 cm line that has been suggested by van de Hulst because I’m going to be detecting it for my doctorate thesis." And he turned around, with a look on his face like, "Go home. You don’t even belong to the radio astronomy club." And just sort of, "Well, that’s very interesting but we do have other things on the schedule this afternoon." [Ewen: Suggest deletion of statement out of context, given the situation] And so that just went down the drain. So I went back home. I thought that was a loss, to get to know all these fellows. Of course, most of the publications were Australian or...

Sullivan

England, yeah.

Ewen

So from that, I then went back and chatted with Ed Purcell about carrying out this experiment and my feeling was that in reviewing the literature was that the Russians really were likely to detect it long before we could get the equipment together. And that was based on Shklovsky's paper.

Sullivan

His ’48 paper, yeah.

Ewen

Right, which was so close on to the time I became involved in looking into things. And it was about the time I’d read van de Hulst and Shklosky and I’d talked to what few in the U.S. who were doing anything in radio astronomy, and it seemed to be one, a difficult experiment, a very risk doctorate type thing since there was considerable questions as to what mechanism we might be dealing with here, or if there was anything we would be dealing with. And so Purcell so then sort of summed it up, since he couldn’t really justify it logically- go to the blackboard and say, "Well, here’s what we..." He justified doing it simply by saying if you can do it then you’ll be in Life Magazine, you’ll never be forgotten and if you don’t do it, it’s only going to cost you a couple of years of your life. And I said, "Fine. On that basis, since we understand each other, I’ll go back and get a better job at the Nuclear Lab for the long pull and we’ll continue to look into this in sort of a hobby way in the interim. Why don’t I take a little harder look at what we know about the hyperfine line because it would be awfully nice to know what frequency very precisely I should build this device on." And that wasn’t a very simple thing to come by in those days.

Sullivan

Well in fact the frequency you were using was measured by Kusch and Purcell.

Ewen

Kusch at Columbia [Sullivan: 1950 publication] and that was the one we used.

Sullivan

Right. But was that experiment done because of your need for a precise frequency or it just happened it was done at the same time?

Ewen

It just happened. It also just happened coincidently that [W. W.] Mumford came up with the gas discharge noise source as a method for measuring the noise figure of microwave receivers just at that time, and that's why I built a gas discharge noise source to calibrate. Ed and I agreed that if we were going to do it, let's do it very right. I think a lot of the care to make sure that it was done right was jointly shared because both of us felt very optimistic when we met, but perhaps a bit pessimistic when we were alone and you want to make sure when you come up with a negative thesis, it is very important that you, for example, have an antenna that you know a lot about. So you think about a horn antenna. It's pretty simple and is well described in the "Rad Lab Series". Although it might have been easier and quicker to throw together a parabola with a feed-at-focus, slot fed British dipole, that was rejected immediately by both of us.

I was really quite pleased in the discussions I had with Ed when I would say, "Why don’t we do this?" or he’d make a similar suggestion, we’d both agree and invariably it was related to making sure that when we said it hadn’t been detected, we knew how far we had gone, how precisely. In other words how much more difficult the task to go beyond the "point".

Sullivan

Or as we would say now, you wanted to make sure you knew what your brightness temperature limit was.

Ewen

Right. We wanted to able to say that this transition is not detectable in interstellar space above this level.

Sullivan

Were you worried about interference?

Ewen

No. Not at all.

Sullivan

At those frequencies in those days, I guess it wasn't too bad.

Ewen

1420 was a very high frequency in those days. Though X band was around, things like that, radar. There wasn’t that much activity going on.

So, we agreed to proceed and that agreement was put into effect in the spring of '49. Decisions then about methods were the point of discussions we had. And I think of the basic decisions were made at one meeting early on. That the antenna would be a horn. That calibration would be by a gas discharge noise source. That the mixer would be a scaled waveguide mixer, based on a drawing in the "Rad. Lab. Series" that had been drawn and placed there by Bob Pound, who was right next door to Purcell's. So if I couldn't understand how you scale that drawing, I could ask Bob. And that we would use a 1N21 crystal, which was a pet also of Bell Laboratories and coincidentally, the crystals were provided by [Harald T.] Friis at Holmdel, at a meeting I had down there drop in at Bell and make sure that I had not overlooked anything in microwaves and putting this receiver together. So the care was just to make sure that we were doing the very best in the state-of-the-art, calibration was appropriate. I recall the afternoon I was ushered into Friis's office, just prior to leaving to go back to Boston. He reached down into a drawer in his desk and said, "I've been waiting for an occasion like this to give some kid the very best crystals I’ve ever seen, the 1N21 crystals. These are beautiful." As a result, I had two of the best crystals from Bell Labs and the Director sent me back on my way to Boston.

And then the Cascode circuit for the preamplifier was a critical item. I then contacted Bill Prichard, who was the Radar Division Manager for Raytheon in Wayland at the time, to determine the status of Cascode or better preamplifiers, developed by Raytheon in their radar effort during the war. He provided me considerable insight on the Cascode amplifier plus some very nice tubes and the latest circuit designs. And Bill, since that time, has went on to become Director of COMSAT and so on. We have kept in close touch over the years, but that's how I first met him was getting into the preamp game through Raytheon. That's why he is mentioned in my thesis. So, that put all the parts together, and from there on in, it was just a matter of gathering together the test equipment and assembling it.

Sullivan

Were these low noise components essential in detecting the line or did they just give you a much better signal-to-noise?

Ewen

The parts that I actually fabricated - designed and fabricated - were the horn, all the waveguide through the mixer, which including a slotted-line prior to the mixer, the gas discharge noise source in full L-Band guide, and all of this was done to be absolutely certain of the calibration more than anything else. For detection, no. [Sullivan: meaning, so much care was not required for merely a detection] I would think, for example, today you could detect 21 centimeters with the electronics in a pack of cigarettes, and a paper clip.

Sullivan

But you could of then also, but you didn't know...

Ewen

But I think I would have been a little reticent though, for example, to place a number on the brightness temperature to the degree I was happy to do then. We were dealing with new techniques at the time but certainly just the idea of what we really mean by antenna beam efficiency for a feed-at-focus parabola is still pretty much out-to-lunch with most electrical engineers even today.

Sullivan

It's not a standard concept.

Ewen

Right. They don't think in that term and I’ve counted numerous problems in that area, just currently. It’s still not put to bed correctly. In answer to your question we took the care to put something together so that, in the words of Purcell, "Faced with a 'wet rock' we will make it as dry as possible." So we’ve got to say this can’t be detected. So that was the care. Now second, you should keep in mind it hadn’t been detected and its presence was very suspect. Even when we did detect it, jumping a little ahead in the story, there was considerable analytical flap over what we detected. We detected hydrogen and we knew we had detected the hyperfine line. Just why it was there and why it had that intensity got to be quite a bit of discussion.

Sullivan

And Ed Purcell told me about some of those questions but let's go back. Another critical aspect was the whole business of what bandwidth and what kind of switching to use. Can you tell me about those decisions?

Ewen

With regard to bandwidth, the thought was that the minimum bandwidth should be somewhere around 5 kHz. This was selected on the basis of considering what spreading there might be caused because of the antenna beam. Not much thought though really about a fully thought out analysis of what we were going to face with the motion of the Earth and the Solar System, that came much later, as a surprise. So, originally it was somewhere in the 5 to 20 kHz bandwidth would be adequate, certainly not less than 5. Second and this in part influence by the fact for a typical AM receiver 4 to 5 kHz is a reasonably easy bandwidth and 20 to 30 gets to be pretty hard to come by unless you are going to modify the receiver. And I decided that all of the processing part, the IF strip, would be an off-the-shelf [Ewen: 30 MHz] communications receiver [Sullivan: National Radio], where you could plug in variable bandwidths and do all that sort of thing. Why discover radio again? So that’s why the construction ended with the Cascode preamp. Then the idea was to second convert, use a 30 MHz communications receiver, and from there on in use the communications receiver.

So first switch frequency, not knowing exactly how wide the line would be, the very first thing I tried was an electric motor with a generator coupled to it, with a Lucite disc that had been coated, half of it metallically, and spin it near the tank circuit of the L.O so that I would get a 30 Hz frequency shift, sin wave. And I managed to get a pulling using a standard National Radio, which was located locally in Medford. So if I had any problems with my IF strip, I’d just hop in my car and go over to Medford and say, "These are the things I want to do." So I mounted that up and I used the generator for the reference to feed my synchronous detector. It was a very simple way to drive the motor and drive the disk and noted that the frequency shift was about 10 kHz. And no line because we were inside the line.

Sullivan

Switching at a rate of 30 Hz, 10 kHz away?

Ewen

Right. 10 kHz was all I could moving a disc up near the tank circuit without completely loading the circuit.

Sullivan

Did you sit down and think once again about motion from the galaxy or turbulence?

Ewen

No.

Sullivan

So you were just sort of move out until you could maybe get it?

Ewen

No, I was looking at it more- the physicist part of my "mountain" was saying to the "valley", "Let's get the frequency right and let's admit that the line width is pretty much unknown and it’s going to be about as gobbledygook as the optical lines or some of the infrared stuff that’s come out." So, let's stay loose on that part of it and not make that a "buried-in-concrete" part of the design. The electrical engineer was saying: "Let's do the best upfront, as far as using an antenna that most people will understand, build it in waveguide, design a mixer based on a unit home-grown at Harvard, well at Rad Lab, Bob Pound, use a Cascode preamp grown over in Raytheon, and design and build this as carefully as possible and hook it up with a communications set that comes from Medford, National Radio, and then proceed to fiddle with things like bandwidth and switching rates at a level of convenience so that at least you don’t"- I knew if I switched 10 MHz, it would probably work but I probably won’t get the job done before 1960. So it was, "Let’s not do anything on the tail end that would embarrass us later, think why’d it take so long."

Sullivan

Now, when you say you did not detect the line when switching 5 kHz or 10 kHz away, did you tune the L.O. over a broad range before you made that decision?

Ewen

Right. The L.O. was turned over a range of about ± 150 kHz.

Sullivan

With Doc Ewen on 12th August 1979. So what range was the L.O. turned over?

Ewen

I believe it was about ± 150 kHz and that was accomplished by cavity-tuning a 3C22 lighthouse tube in a radar jammer cavity. It was a surplus World War II device. It was very stable.

Sullivan

What was your form of output?

Ewen

The method of determining the frequency was quite linear with just a manual drive on the cavity for that small of range and the method of measuring frequency was by that very complex, in those days, synthesizing harmonics with General Radio equipment that was strewn all most over the laboratory. Most of what I had in there in that laboratory was frequency measuring equipment, and we would count up, then down, and check, and check. And this was then evaluated independently for me by... [Ewen: Most of that equipment was provided on "weekend loan" by the Nuclear Lab. I would bring it over to Jefferson in a wheel barrel on Friday afternoon, and then back to Nuclear the first thing on Monday morning. Weekend holidays were cherished. Our payoff weekend was Easter. [Sullivan: not on tape. 25 Mar 51 = Easter Sunday]]

Sullivan

And your signal output was just a strip chart?

Ewen

That's right. Just an Esterline-Angus strip chart, crank it up with a crank.

Sullivan

I've seen pictures, where you are there with headphones on. Was that just testing equipment, or were you actually doing observations that way?

Ewen

Well, occasionally I would put on headphones when observing in the evening just to listen to the heterodyne beat as that oscillator was being tuned and I had an old comfort chair brought in and put in the lab. And I would just close my eyes, and snooze a little, and I would listen to that thing whistle and if it gurgled or popped, it was almost like babysitting. I’d be able to come to. And I would also tell the night janitor what times to definitely wake me up. But I could tell that gadget when it was running correctly, it had, because of the number of gears and motors, peculiar little noises coming out of it. I could walk into the room and I’d just listen and I could tell you whether the receiver was working or something had happened.

Sullivan

Let's go back to your attempts to detect the line. So 10 kHz did not work. So, what time are we talking relative to final detection?

Ewen

It was probably in the late summer, fall of ’49 or '50. No, we didn’t really get it together until early ’50. [Ewen: late ‘50s, just before Christmas]. Then the next step was to separate the frequencies further. No other changes were made and to accomplish that, it then became necessary to introduce two oscillators in the communications receiver. That was not a large modification. I also at that time did by just simple interchange in the communications receiver, change the bandwidth to 15 kHz, to improve sensitivity a little, wondering if I was really improving the signal-to-noise, so I didn’t know whether I’d gone out of the line for sure. So, then we tried both a 5 and a 15...

Sullivan

But since you had been switching to 10 kHz anyway, that probably told you that you were inside the line.

Ewen

No, I was still working on [Sullivan: the assumption of] the negative thesis.

Sullivan

That’s what I’m saying but, by broadening the bandwidth, you couldn’t be sure that you didn't have a super narrow line.

Ewen

The whole focus of attention was to separate the frequencies, but that would be the only thing which would be of any value in searching for the totally negative thesis.

Sullivan

But now in fact, you just said you were not worrying about Earth Doppler motion. But if there had been a very narrow line, you would have missed it, would you not?

Ewen

Absolutely. [Ewen: narrowing the bandwidth by a significant amount would degrade sensitivity and have an adverse impact on frequency stability and tuning rate.]

Sullivan

It was by luck it had been right at the line.

Ewen

As a matter of fact, it was detected by luck and I’ll tell you how.

Sullivan

Well, we are getting there.

Ewen

We are getting there. I then put the two oscillator system together but it was fixed. I was unable to adjust it, do it quickly. Switching then was something like 25 kHz when it was on square wave, a little neater job than this wheel. And nothing that I could really say was a detection. Some little gurgles and burbles but they could have been noises that the machine was generating for itself.

So then, we had a summit meeting with Ed. As I recall it was in the late fall of '50 and it was, "Are we ready to go for the negative thesis?"

And Purcell was just a giant at those meetings. Rather than discuss whether it was time to go for a negative result, he was more upbeat on, "Is there was anything else that you would like to buy or try before you let it go down the drain?" And I said, "Well, it would be very expensive?" And there was just one other thing I could try in order to separate the frequencies even further but to do it I would have to buy a new receiver. There was a new one which National Company had just come out with and it cost $300. I would modify that and if that didn’t work, nothing would work. But we don’t have that kind of money because we only had $500 for the program, and we’d spent the $500 on the horn. So he said, "Well, if that is your decision, come on back tomorrow and we’ll finalize it."

So I came in the next day and I’d mentioned it to him occasionally, and this is one of the things that he has never given me an answer as to how this happened. But he said: "Now you‘re sure. $300 and you buy that receiver and that’s going to be it?" He reached and pulled out his wallet and took out $300. I don't know where the $300 came from. It was cash.

Sullivan

Laundered.

Ewen

So there it was. I said, "Yes sir. I’ll do my thing." And I bought and modified the receiver and did that with loving care, working with that receiver. And the very first night I turned it on, put that thing on, the line was detected, very peculiar way.

Sullivan

What was your switching now? How far away?

Ewen

75 kHz. And I knew that everything sounded good. The machine was really running beautifully. I was scanning the ± 150. And I started scanning through the rest frequency and that was just a big nothing. All the way to the end and I thought, "Oh well, just let it run because maybe Columbia goofed [Sullivan: in the lab measurement of rest frequency]. "And at the very end, and that's probably the record I should have kept- at the very end of that strip chart record and this was about 6 feet long and at the right angles, you see this things go up and just again it went off scale. I said, "My god, what could that be? That is too resonant for anything I’ve got in this machine." So my feeling was 100% that it was an equipment error that I was unfamiliar with, some sort of resonance.

Sullivan

This was 75 kHz off the rest frequency?

Ewen

No, 75 kHz switching and 150 kHz tuning.

Sullivan

Oh, 150 kHz off.

Ewen

I was at the end of the scan. I couldn't go any further. The oscillator wouldn't scan any further. The motor boom, hits the limit and that’s it. So I just moved everything back 150 kHz, cut the gain a little, and the then goes off the chart.

Sullivan

But you had enough confidence in your receiver though?

Ewen

Oh, yeah. I didn't think it was the line, but I knew that...

Sullivan

That you could investigate it anyway...

Ewen

That I could quickly investigate it by shifting over here and scan that part. But had I stopped at, say, 100 kHz instead of 150 kHz, I never would have seen it.

Sullivan

You would have had your negative thesis.

Ewen

That's right. But the minute I saw that- what the devil could be 150 kHz. And that’s what I was thinking sketching this thing out. And that's when I called the Harvard Observatory. It was about 2 o'clock in the morning. And I said: "Physics Department at Harvard and we’re performing an experiment here."

Sullivan

You got someone at 2am?

Ewen

Yeah, at 2am, some grad student on the phone. And I said, "We’re over here in the Physics Department and we are performing an experiment that involves astronomical coordinates. And I can tell you that we are now looking almost directly south at a declination of -5°. But what I’d like to know is, can you tell me the component of the velocity vector in that direction as a function of hour angle, say from midnight to three in the morning because I'm trying to measure a Doppler shift?" There was a big pause, "Who are you?""Physics Department - Harvard." "What are you looking at?" "Interstellar gas. I'm looking at a resonance in interstellar gas and I need a Doppler component from you fellows." "Well you better call us some other time. Doppler, what are you guys doing over there? Is this a prank call?" So I said, "I’ll get in touch with you later." Just zip. After that I didn’t call the observatory. I just got the books and looked it up and computed it myself.

Sullivan

There had been work done on that sort of thing before.

Ewen

Oh, sure. It isn’t that complicated but at 2 in the morning, you figure-heck, I’m sitting right here at Cambridge, why can't I just call the Harvard Observatory and have a fellow just spew that to me over the phone. Why should I go to all the details of looking it up?

Sullivan

In fact, did you worry at all...

Ewen

See, I hadn’t worried about anything.

Sullivan

No, but about where you were looking in the sky or were you just tune on when you got things working?

Ewen

The only thing that I was concerned about was to pay attention when the Milky Way went by.

Sullivan

OK, so you did do that?

Ewen

Right. That was the only thing I worried about. I knew I was looking south, I knew when the Milky Way went by and line came in just about ballpark.

Sullivan

But you didn’t care whether it was daytime or nighttime, for instance?

Ewen

No. The only thing that made nighttime observing attractive was that’s when I wasn’t working at the Nuclear Lab. So in the daytime the machine ran by itself. I had it set it up so that at noon I would go over, crank it on because it’d be just 12 hours later and the Milky Way was going by around 2 o’clock. So, I was just sketching the negative side and I’d just crank around and let her do her scan.

And then, it would automatically stop at the end of the scan. Then at night I'd take a look at it and I remember after I got this detection, I was picking up that one on the backside of the Galaxy. See, the thought was that the whole astronomical input at that time was that if you're going to see anything, you're going to see it near Sagittarius. You need a big, dense cloud and that fellow is right in the middle of the galaxy. This idea of looking around and casually seeing hydrogen all over the place hadn't occurred to anyone. When I saw it in the anti-center, that thing came in really strong and narrow. And I just dismissed that. I was too busy getting it over here at night where it belonged at night when I was there. If it wanted to do it in the afternoon on its own time, I would just take those and pile them up, and forget that.

Sullivan

This is now a few months later?

Ewen

No, this is in the same time as the line was being detected because it was picking it up in the afternoon for me automatically. But I wouldn’t believe it because I would figure that was the day of "MIT’s on the air," somebody is goofing me up, the car...

Sullivan

No mention about that in your thesis at all?

Ewen

No, I never mentioned- I never mentioned it to Purcell until much later, that I had the anti-center, that it showed up over there That was even neater and clearer but I wouldn’t believe it if I wasn’t there.

Sullivan

This brings up a related question. How was the angle of the horn chosen? In other words, declination -5°?

Ewen

It was chosen on the basis of the waveguide bend that was commercially available at L Band, that would allow me to bring the waveguide horizontally out the window and then up. That comes out at -5°.

Sullivan

Did you pick a room that was facing south purposely?

Ewen

Yes, I did. I did do that, definitely wanted to look south. I wanted to be on the local meridian. I did check out the road that we were within 10° or something. But that reminds me of one that Friis taught me because I asked him at one time, "Why were all antennas that Kennedy was building and were being used in tropo-scatter, why were those 28 feet? And why weren’t they 30 or some number of meters?" And he said, "It turns out that plywood is 14 feet long," and he built the first one as a model in wood and then fellows came along with their aluminum- 28 feet.

Sullivan

Well, that’s how Grote Reber’s- his was a 30 foot dish for the same reason, the length of wood he could get at the local lumber store.

Ewen

And that’s what Friis was doing at Bell.

Sullivan

It may be the same lumber standards that made that 20 feet the same as Reber’s 30 foot dish.

Ewen

But somebody put a vertex in the middle for Reber.

Sullivan

Ok, I wanted to ask, frequency switching. Where did the idea of frequency switching come for that?

Ewen

The idea was at that initial meeting with Purcell and I think was just a natural for nuclear resonance. And Purcell said, "Just frequency switch."

Sullivan

Was it a standard technique at that time?

Ewen

All of the other grad students at that time were doing nuclear resonance type work, and at that was fairly standard. As a matter of fact, trying to decide what we would call a thesis and what discipline was the thesis- my final doctorate sheepskin says "Microwave Spectroscopy," so trying to decide what hill we were going to be on. We are going to be in microwave spectroscopy and that is the standard technique.

Well, it was Purcell who suggested it and I think it was unfortunate that he didn't get greater recognition, particularly that that idea wasn't recognized as a rather significant thought because Dicke’s contribution was really "switched load". The "switched frequency" mode is not really a true Dicke. So that was a whole new radiometer type, much the same as the [Whiteoak?] machine or whatever. But it has always been called the Dicke radiometer.

Sullivan

But was he the one who developed frequency switching for NMR [nuclear magnetic resonance].

Ewen

Purcell, I can say without question was the one who suggested it for the radio astronomy application.

Sullivan

And for NMR also, is that correct?

Ewen

I believe so.

Sullivan

You were working all the time on the accelerator? So, this was really, like you said, a hobby?

Ewen

This was nighttimes and weekends. I was putting in 40 hours at Nuclear.

Sullivan

And Purcell, also, you said he was away a lot of the time.

Ewen

He was tied into those committee meetings for the Korean War and he was fairly high up with the Defense Department. So most of the fellows were shuttle-hopping to Washington, most of the key scientific types in those days. And I think he was essentially assigned almost on a sabbatical for about a year assignment in Washington. But at the times I needed him he was there. He participated in all major decisions. And certainly while we’re hacking wood and putting things together- in many ways I benefited possibly by the fact that he wasn't always there, because I could get some things done at the Nuclear Lab and I wouldn't have to say: "I haven't done anything for two weeks." But I knew what the schedule was, and they were all put together and he was a very good advisor in the sense that he never really bugged me. He wouldn’t even call and I would wonder...

Sullivan

Whether he cared at all.

Ewen

No. I would look at the phone and say: "Boy, if he calls, what will I say?"

Sullivan

So, having this initial detection which looked pretty good, you said you now thought about the whole business of Doppler and within a few days...

Ewen

Oh yes. Almost immediately. As a matter of fact when I first discussed it with Ed, I recall the night it was first detected. It was 25 March and within 2 or 3 days right around there, I had it really nailed down and understood the frequency shift. And then I just picked a morning when I was confident that that was it and everything appeared reasonable, that I recall taking the Esterline-Angus roll of about 30 feet and I knew where Ed would be coming in in the morning, he usually came down a long hall in Cruft. And I had my old, bloodshot, red eyes and I took the roll and zipped it about 30 feet, right at his feet and said: "There it is." And he saw me, "Really?" There it is. It was just beautiful. So from there on it, it was, "Well, come on in the office and we will find out what had been detected."

Sullivan

Well now, there still was there not some luck involved because you calculated the orbit of the earth and got that motion and so forth.

Ewen

Well, I got Hercules too. I knew the Sun, the Solar System...

Sullivan

You got that, the whole Galaxy. But, if the line had been at -100 [Sullivan: Km/s] with respect to the local standard of rest, you still could have been way off.

Ewen

No. I was strictly assuming hydrogen at rest and the Solar System was the only one moving in the whole show.

Sullivan

So now, what did you do at this stage? Did you follow this up or confirm further?

Ewen

No. At this point, it was treated as an experiment to be performed in physics. It had been performed. It’s complete. If it had been performed by the Astronomy Department, it would have been a different thing. But in the Physics Department it was strictly an experiment in Microwave Spectroscopy, very convenient laboratory. It worked and there was obviously some significance that would be drawn by astronomers. So it was in that context that we wrapped it all up. It was at about that time that Purcell then said, "For the publication of this, in the field of radio astronomy, who are the people, because I’d done that literature search, who are the people and who do we get in touch with?" And then he sent a wire to [Joseph L.] Pawsey in Australia and one off to- well, it was going to be the Netherlands. I don't know the details of how he discovered that van de Hulst was at the [Harvard] Observatory. He may have met him, I don't know. It was not my opinion that he had. I have no recollection of any thought that he had met him. He certainly never discussed it with me other than it was coincidental that van de Hulst was there. And now that we had gone through the two months [Ewen: more likely weeks] of assuring ourselves that we had detected the line, he said, "Why don’t you go and see van de Hulst." That was very much like Purcell, who though a very important cog in the whole wheel, it was typically of Purcell to say, "Hey kid, you did it, you tell him." And which to me, I looked at it that way, my eyes cloud a little, that’s my father saying that, you know, "Hey kid, be a big guy. Go tell him." So, I went up to the Observatory looking for Hank van de Hulst, located him, and just very quietly told him my name and what I had done with Ed Purcell. And he was quite shocked, to say the least.

Sullivan

He had known nothing about this operation at all?

Ewen

No. I had never talked to him before. He was quite taken aback about all this having happened. And so then he asked me for details of where we were looking and how it was done. I had brought along some strip charts, but I hadn't brought schematic drawings and those details. But he then placed a call immediately to [Jan Hendrik] Oort, and when he got Oort and [Christiaan Alexander "Lex"] Muller on the phone.

Sullivan

While you were still in the office?

Ewen

Immediately. And he then described to them where the detection had occurred and what the intensity was. And then they asked that I get on the phone and describe how it had been done, how "switch frequency" worked, what the noise figure that had been measured, what the signal/noise ratio was and how the overall machine was calibrated and the use of this noise source- things like that, that apparently were new to the Dutch. And so I went through quite a discussion in some detail and agreed then that I would then send schematic drawings and any parts, like 6AK5 low noise preamp tubes, that weren't available conveniently to the Dutch, to get all of these to van de Hulst so that he could expedite the shipment to Holland. And then the understanding was that we would appreciate confirmation, since they had proposed the detection. I did not then meet again with van de Hulst until the time of my doctorate oral. He was on the committee with Purcell of course and some others.

Sullivan

When was that?

Ewen

That was probably a couple months later and the interim was occupied by the fact that I had completed then the thesis and I wrote it up. I got up to 47 pages and then take the oral. But I had one other thing that I had not completed and that was my German qualification. So I packed a bag and went to Vassar, where I knew a friend who spoke German, and I spent, right after the detection, right after speaking to van de Hulst, 2 or 3 weeks out at Vassar studying German. And that got to be pretty well known by Bok, who was in South Africa at the time. He said Oort was furious. How could anyone detect an interstellar line and then be studying German? Because that’s what I was doing, because otherwise I wouldn’t get my Ph.D. And I remember getting the German out of the way. The professor in charge of the German exam was Oldenberg, atomic physics, and Oldenberg selected a chapter out of [Oppenheimer?] and Becker on magnetic theory. And he told me about the chapter. He said, "Here’s the book. It’s in German so I’m sure you are familiar with the text. Just translate this chapter." And he left me in a room, all by myself, and through the door came [?] and if you know [?], who is the seat of the chair of Einstein, he’s quite a speaker and a mother of grad student or possible undergrad student at Trent, [?] was right there at the door of the room where I was trying to get my German out of the way. I’d spent three weeks at Vassar. And [?] was going, with his hands going up and down, telling about this poor kid was going to have to be kicked out of school for awhile. So I just picked up everything and went up stairs to another room and then Oldenberg found me up there. He said, "My god, thank heavens I’ve found you." "You know what you’ve done," he said. "You’ve violated this and that. I should notify the Dean." So I told him about [?] and I said, "You can ask him. He was blowing my mind down there. I just went to another empty classroom." And that’s all that happened. So then he quickly looked at what I’d written and he said, "That’s a beautiful job of the chapter but it’s not the edition you reading." But he said, "I think you know German anyway, know a few words." And that was it. So that’s that aside. And then as I said, I didn’t see van de Hulst again until we got down to oral time.

Sullivan

Now just let me check on dates here, because the Dutch detection was in late April around April 20th. [Sullivan: Wrong, 11 May] So, you must have spent only a week or two in your confirmatory observations before you went and told van de Hulst.

Ewen

We told van del Hulst... I guess we... I may have that wrong. I see what you mean. My feeling at the time- I had wrapped it up in three days- was that I had it. The great concern of Purcell was that I the MTA or a high line... [Sullivan: like the MTA!] [Ewen: Last note in lab notebooks re observations is April 22- then it begins again in the fall (’51).]

Sullivan

That’s right, you haven't said that for the record. There were some interference possibilities.

Ewen

When I showed Ed the chart and said, "There it is," he said, "Come to the office. We’ll see what you detected." And I was a little concerned when he got on the phone and called [Sullivan: MIT] Tech and wanted to know what they were doing. He was interested that there was something on the paper but he was then going to search out what it was. And I was quite confident because everything checked, but it had only been 2 or 3 days that I’d worked it. And it may have been at that time and that was certainly the motivation for confirmation with Ed, to not only how to you join the radio astronomy club but how do you approximately make certain that this has been detected because though you can try a lot of things individual- he did recognize the significance of the measurement as an interesting experiment in physics...

Sullivan

He wanted independent...

Ewen

So he may have been motivated very early on then to say, "Go find van de Hulst." Because I know even at those early meeting, he was discussing who should we get in touch with and settled on the Dutch and the Australians. And though the English were quite active at the time, we did not contact them.

Sullivan

Do you remember any advice from van de Hulst on how you might- from an astrophysical point of view - improve your technique or follow it up or specifically if you might go to a more extended frequency switching?

Ewen

No. The first discussion that I had with van de HuIst was the original one after the line was detected and then at the time of the doctorate oral, when after I had been asked a couple of relatively simple questions in physics, the meeting broke up with Purcell and van de Hulst at the blackboard for about two hours to discuss what this all meant from the physics and astronomy standpoint. And I can recall sitting there, dozing, and listening to those giants with my feet up on the chair until at some point in time Purcell looked around and said: "Oh, you're all set Doc, you can run along."

Sullivan

You thought this was still part of your exam?

Ewen

Right. I was waiting for them to turn around and say, "Now that you’ve listened to us, what do you think?" But fortunately, he said I could leave and so that was the only other time. And the next time that I met van de Hulst was at Leiden. I took a trip over there sometime.

Sullivan

It has always interested me that those two papers in Nature that came out are so different in their approach. Your paper with Purcell is physics of the line, while their paper is entirely galactic structure. Just an different approach.

Ewen

And the telegram from Australia - detection!

Sullivan

Right, we’ve slapped together a receiver. We’ve got it.

Ewen

It works.

Sullivan

So, you got your degree right only three months after the line detection and then was there no more follow-up in the Physics Department? Purcell said, "The experiment is done."

Ewen

None. The experiment was done and we notified the Harvard Observatory, of course, because I had been there with van de Hulst. Harlow Shapley was the Director at the time and there was no real interest to pursue it further. No interest at all at Harvard. So we just packed up shop. There were some visiting firemen who would drop by occasionally. [Sullivan: Taffy Bowen was one of these (letter 10/22/51 from Purcell to Kerr)] And I’d be over at the Nuclear Lab and Ed would call and say, "So and so is coming into town and he wanted to see the machine." So we would converge there for a couple of hours, turn it on, make it do it’s little Rube Goldberg audio thing, and if he happened to be there when the Galaxy was around, we’d get a line that he could take home. So that was about it. It was a little pony show. And I was then at that time very involved with this Korean War thing and being called back to active duty and problems with whether I was going to be in the Navy and finish up. There was a lot of confusion in that area so that as far as operating the equipment any further, no.

Sullivan

So basically you were working on the accelerator during those years?

Ewen

Right, getting the proton beam out of the cyclotron and that occurred just three weeks prior to the detection of the line.

Sullivan

Oh. I see.

Ewen

It was just bang-bang and everything was over. We’ll just wrap up and decide what comes next and try not to have another war.

[Ewen: The time period between the detection of the line on March 25th and graduation in June was a bit hectic. First, came the oral followed by two weeks of studying German and my qualifications exam in that subject. One week to write the thesis. Then, "Greetings" the Navy ordered me back to duty. Apparently, they felt I would be more useful as an officer in the Korean War than a graduate student at the Harvard Business School that fall. I had applied and had been accepted just a few weeks earlier. I was ordered by the Navy to report to Washington, D.C. for a two week period during which my "Q clearance" was being processed. Following the receipt of the "Q clearance", I was scheduled for assignment to the Navy Nuclear Group at Sandia, New Mexico. During my two-week stay in Washington, I submitted an unsolicited proposal to ONR describing a unique technique for the passive underwater detection of submarines. I was awarded a $50K contract, released from active duty and returned to Boston to act as principal investigator on the underwater warfare contract, which I had arranged would be awarded to the Scientific Specialties Corporation. At that time it was headed by a former director of the Nuclear Laboratory. The kick-off meeting for that contract was scheduled the same day as commencement, as a result I missed the opportunity to wear the doctorate robe. During the summer of '51, I spent most of the time on a 65 foot yacht cruising around Boston Harbor and the adjacent islands experimenting with my underwater detection scheme. Other than occasional visits to Harvard, most of my time was fully occupied for the rest of that year with the underwater warfare technique (which was quite successful, by the way). [Sullivan: Not on tape]]

Sullivan

You were in a race on the accelerator too.

Ewen

Yes. That was with [Enrico] Fermi at Chicago which I felt was the great accomplishment. If I were asked at that time, "What is the greatest thing you did in ’51?" And certainly for the next two or three years, I would have answered, "The external beam from the Harvard cyclotron. [???] the cyclotron, get the external beam, and beat the boys in Chicago." And that was it, that’s where I spent all my time.

Sullivan

It would be interesting- Let me ask you, at what stage did you feel that detection of the 21 cm line was more important?

Ewen

Well, it started about a year later when Bart Bok came back from South Africa and hat was also the beginning of Ewen-Knight. It started with Bok and when he looked me up when in hours of when he got back into town. And at that time, I was working on the underwater warfare problem, which was more important than my being in Korea and I was working for a company called Scientific Specialties, which was an outgrowth of the Nuclear Lab. And Bart looked me up and asked what the status of the machine was. I believed prior to his return- this is the part that Ed and I haven’t quite resolved how it happened- but that machine filled a room about the size of this room, just giant, and in that period about the early summer, during the summer...

Sullivan

Of ’52?

Ewen

Of ’51.

Sullivan

Oh, ’51. That’s right. But Bok came back when now?

Ewen

Bok came back, I believe, early ’52 or something like that, somewhere in there. About a year, it was probably spring on ’52. It was in between those two times when I was working during the summer, I remember a call to go to Washington to talk to Vannevar Bush, and we walked in a park in Washington. He told me the great virtues of being a fellow at Carnegie and that they had decided that I should come to Carnegie and be a fellow and I wouldn’t have to worry about anything the rest of my life, which immediately convinced me I didn’t want to go to Carnegie. He said the wrong thing. So Merle Tuve I guess had sort of put him up to it, the head of the Department of Terrestrial Magnetism. But Van Bush and the boys at MIT- I think the president of MIT went to Caltech, who was he? I got to know him...

Sullivan

At that time?

Ewen

Right, but he was a buddy of Conant’s and I believe he was president of Caltech.

Sullivan

DuBridge?

Ewen

DuBridge, right. DuBridge and Van Bush were buddies and of course, DuBridge...

Sullivan

He had been head of Rad Lab?

Ewen

Right, with Purcell and everyone- It was a real tight club. So between DuBridge and Van Bush and all, I don’t know what part Purcell played in all of those because in those months, he was back in Washington and I really didn’t get to see that much of him other that an occasional call of, "When are you coming down?" But I remember describing this whole thing, how to operate the machine to Merle Tuve. It’s my understanding that an arrangement had been made to ship all of it to Washington and I was on my way back to my submarines. So a truck arrived one day and we put everything in the truck and it went to Washington. They didn’t want the horn. That went to the Smithsonian and then went down to Green Bank. And so, Merle Tuve and Carnegie had the original machine. And it was much later, I think it was in the late fall of ’51, about in there, I had a call from Carnegie. They’d put it all together, dusted it off, and it didn’t work.

Sullivan

Of course.

Ewen

This is what I call a confirmation. So I buzzed down to Washington and fiddled with the machine and they’d put it all together, there wasn’t much fiddling. And they had a big 60 foot antenna I believe, right there in downtown Washington and pointed it off toward the local cloud and in it came. And Tuve made an interesting comment at the time. He said, "When you sit it down,"- part of it I’d built out of 2x4s because we couldn’t afford relay racks, things were sitting on little shelves there, so they had all these little wooden things and I had a placed to hand my jacket, a little hat hook on a 2x4, he said, "You know Doc, we saw that and we put all that together and we put your hat hook up." He said, "We didn’t realize the things doesn’t work unless your hat’s hanging up." I said, "That’s it. Stick the hat on and see if it works." It was one of the those things. Then, with Bok when he came back, he looked me and he apparently raised the roof with the Harvard Observatory, with Menzel and Shapley and how ridiculous this all was.

Sullivan

Letting this all slip away?

Ewen

Oh yes, just absolutely ridiculous. Not only letting the equipment go, but letting this team fall apart. Well if you know Bart, you know he really put it to them.

Sullivan

I’ve talked to him.

Ewen

He did it all. At that point, he just said: "End of nonsense". He pulled the team together. Got hold of Purcell and myself, Shapley and Menzel and we all went over to the Harvard Faculty Club, [???]. We had a little summit meeting to decide what Harvard was going to do about this, to pacify Bok because Bok was going to do something about it. Very quickly it became apparent that Bok had convinced Shapley, and Shapley was ready to go and Harvard was going to move into the field of 21 centimeter research. It would be a major effort. We would immediately go for NSF [National Science Foundation] funding and put Harvard on the map. This was the chance which all U.S. radio astronomy had been waiting for, to get something going here, rather than it running only in England and Australia and so forth. This would be the focus. And so, of course, with Bok, you just nod your head. So we nodded and then Shapley said, "Well, now we have to figure out how to do it." And so he turned to Purcell, "How are we going to do it? Can Doc build a machine for the Observatory?" I said, "The only problem is that I’m working and I’ve got this thing with Scientific Specialties. How about getting the Engineering Department to do it?" And Purcell said, "Engineering Department? Did you have anyone in mind? Harvard or MIT?" "Well, we’re talking about a Harvard program here." "We don’t have an Engineering Department so that is out of the question." "How about MIT?" "No, not them either, it’s Harvard"- because while MIT and Harvard are close, they aren’t all that close when it comes to something like this. I was picking up all these little niceties at the table. Finally Purcell, and this strictly a Purecellism. He said, "I’ll tell you what we will do." He said, "Doc, you know how to build it. We can’t build it in the university. Why don’t you form a company, build it, and then liquidate the company and that way we will have the machine." And he said, "Now all I got to do it figure out how much it will cost and Harlow, you just pay Bok. So it will be his corporation. So it will be Harvard University, some corporation, build it." And this was Purcell, my business school manager, and Shapely said, "Well, it sounds great to me." Casually Bok said, "As long as we get the job done." So here I was sitting with these fellows and they all start looking at my end of the table saying, "Well, what’s the name of the company and when are you going to have it ready and how much is it going to cost?" And I said, "Wait a minute." Purcell throws an idea on the table and suddenly everybody is jumping me for answers. So I said, "I’ll be back. I’ll get in touch with you. See whether we can do something." So I got in touch with my buddy Bill Pritchard and we spent an evening going over- Bill is at Raytheon- what it would take. And maybe could Raytheon build it, let me go back to my submarines and what we came up with probably cost around $100,000 in those days.

Sullivan

Right, that’s a lot.

Ewen

And no way we were going to get that kind of money for hardware into a company out of NSF. We could get couple grad students on it but not that. So we fiddled and fiddled with it and finally just the bare pieces for the receiver were going to cost somewhere around $10 or $15,000 and Raytheon wouldn’t touch it. Pritchard told me, "They wouldn’t let it in the door." So then, the problem was looks like I have to form the company and I had just about wrapped up my project and the Korean situation had cooled enough so I thought I could afford to go into astronomy without making it look like I wasn’t needed at Sandia or something, it keep clearance- because that’s what I wanted to be thought, to live out my life at Sandia making bombs. So then the decision was made and I got back with Shapley to proceed, form a company- Purcell, at that time, said he was not going to, I had to work with the Astronomy department- and that it would cost $15,000, take about six to nine months. I’d form a company. I’d need a $1,000 to capitalize the company so I borrowed that from a guy by the name of Knight, who was an MIT graduate and was also an Amherst graduate- an old buddy of mine. And Geoff Knight, who did a thesis on [quadripolar?] moments at Tech, put up $1,000. So we didn’t waste time figuring out what we would call it, we called it Ewen-Knight. He’d go back and do his thing because he wasn’t in radio astronomy, he was in having a lot more money available instantly than I did. And two years later, I paid him back his $1,000 but retained his name as Ewen-Knight, which is spelled with a K. That’s why we spelled dae, dae, because we didn’t spell Knight right. In other words, Harlow Shapley’s best statement, he said, "When you meet Doc Ewen, you’ll find out he’s keeping everybody in dark about Knight." Yeah, that’s Harlow. And so, we formed a company and located it at 698 Massachusetts Avenue, which is right at Central Avenue, in a furniture store, second floor, short term lease, build it and get out.

Sullivan

That was going to be it?

Ewen

That was going to be it. Didn’t hire any full time people, just moonlight Harvard, MIT fellows at night, the best technicians they had, mostly pulling fellows out of the lab down on [?] Street, the Instrumentation Lab.

Sullivan

The Instrumentation Lab.

Ewen

Right, that’s where I got most of the shop stuff. So we go to work at 6 o’clock at night. We’d work until 3 in the morning and, then I’d go back to work, sort of red-eyed, which is kind of the way we do it now with the troops.

Sullivan

But let me ask though. Undoubtedly you could have had a position at the Observatory yourself, if you wanted it? Obviously you chose to go this route.

Ewen

I did. That was one of Bok’s- I failed to mention it. At that meeting, Bok insisted that not only would we implement this engineering wise, as Purcell suggested, but that additionally I would join the staff and teach the engineering...

Ewen

I did. At the meeting at the Faculty Club, Bok insisted that not only would we implement this engineering wise, as Purcell suggested, but that additionally I would join the staff and teach the engineering aspects of radio astronomy to the grad students.

Sullivan

Oh, you did have a position on the Harvard staff?

Ewen

Yes, I was immediately appointed to the staff, and jointly with Bart teaching the course in radio astronomy, which was the first course taught in radio astronomy anywhere in the U.S.

Sullivan

Well hold it now, to be fair Van de Hulst, he gave a two semester course.

Ewen

Oh, you're right. I’m talking about leading to... [Ewen: a Ph.D. in radio astronomy.]

Sullivan

So, it's a little bit different than I was thinking then. I was thinking that you were off with your company and Bok was at Harvard, but really you were both actually at Harvard.

Ewen

That's right.

Sullivan

How long did that sort of continue?

Ewen

As long as Bok was there. Right up until he left.

Sullivan

’56?

Ewen

Right and it was shortly after, I guess, [Thomas] Gold left and I recommended to Bok that Lilley be brought on from Yale. And after Lilley had accepted and moved in and Bok went to Australia, I left Harvard, actively teaching.

Sullivan

Right but you still supplied receivers and...

Ewen

Well, not really that much. Receivers for academia are a place to really put yourself out of it [?], that’s not really a good thing. My problem was I didn’t want to be in business in the first place. There was another question, how this lasted so long because I delivered a receiver. But we were about half into it and John Hagen called from NRL and said they wanted one. I didn’t occur to me at the time the reason he wanted one, of course, was that if he had one and these graduate student- which was really what I meant by the first course, it was a graduate course leading to doctorates and the students were the Drakes and the Heeschens and the Lilleys. They were all students we were producing and many of the places they were going, the instrumentation wasn’t available. And when Hagen called, I thought, "Well, we’ve got about three months into the Harvard machine. At least we know enough about our mistakes, if another 15K- so it was a 15K type thing. I would build two, one for NRL and one for Harvard and that will be it." And I was saying that all the way through the one for Berkeley, and the one for British Columbia, Michigan...

Sullivan

DTM?

Ewen

Yeah. Everybody wanted one. And then Green Bank, I think about the last one was for the National Radio Astronomy Observatory. And they said Doc, you got to do it because many of these fellows thought the contribution I’d made was getting the line, that’s one thing, but then putting the United States in the business of getting data and cranking out doctorates, showing that an astronomer and a [???] could work together to produce hardware and doctorates and put the U.S. in the business almost over night

Sullivan

Now by the time you made this last receiver for NRAO, of course, Ewen-Knight had gotten into many other things.

Ewen

Right.

Sullivan

It was just a very minor part of the...

Ewen

Yeah, the problem was that there were so many of these going along and then I had to make some pretty basic decision whether I was going to stay on at Harvard and that was about the time that Bok was making some basic decisions whether he was going to stay. I had enough interfacing with the academic life to have some question in my mind whether that was really the life. So when he left, I pretty much made up my mind and bit the bullet and gone in other directions so that- Actually at the time he left, Ewen-Knight was very much into things like tropo-scatter. We build the 22 tons of transmitters that went on the Texas Towers and things like that. Doing completely different things and that’s was I formed E-Dae at one point was in case there was a need for these receivers, to have them done separately in a way that I could keep track of my losses. That’s my friendship club and that was wise and that’s their thing.

Sullivan

Now, I hadn’t heard about this before, this E-Dae company. Was that just a play on Knight, Dae?

Ewen

That’s right. It also was the initials of- ’56, Mary Ann and I were married. So we decided our first born would either Donald Arthur or Debra Ann. So we had DAE and we had the initials and said, "Ok, how about Ewen-Dae?" Very shortly, I plan to liquidate both of those corporations because I’m now very much in real estate as well. They are in too many things. I’m trying to condense it all into another company and that company, I would think, very shortly is going to be on the books. I’ll be using the initials of my fourth son and he’s Mark Edward Ewen. And I thought the Ewen-Mee and don’t spell "me" right, with two Es. So I assure you within a couple of years or maybe this year, we will have a Ewen-Mee Corporation. So Knight, Dae, and Mee and we never sell any of this right but...

Sullivan

Now, getting back to the mid '50's. I agree with you on your assessment of getting radio astronomy going in the U.S., but what would be your comments on the science? I mean, certainly the people it produced were critical for the U.S. in the late 50's and 60's and even today, but what about the observations that were done themselves at Harvard. It seems to me that there was a loss because these people were not building their own receivers as was the mode of radio astronomy up until then. There was over-interpretation of bumps and so forth, which lead to some well-known misinterpretations. I mean, do you agree with me on this and if so, how did this come about?

Ewen

Well, the question is: Should the observer be more than intimately familiar with the design? Should he be a hands-on type? I think the students knew enough about the instrument. I would in the long term haul and I made a comment at the time that I felt ultimately we should be able to hook a computer up to an antenna, and let it operate remotely somewhere and provide the information for astrophysicists to interpret. And the assurance that the data was valid should be provided by engineers on-site who maintain and operate the equipment. But other than specifying what is to be measured and then transferring that to the other mountain- I would leave responsibility for instrumentation, calibration, and operation with engineering. I would insist that the one that is to publish the paper and carry out the research be responsible for all the pointing programs. In other words, where are we going to look? For what period of time? What will be the parameters?, etc.

Sullivan

Not the calibration also?

Ewen

Calibration as well. Calibration is external. The fellow should think about the "box" that is going to do something for him and he should never have the calibration in that box.

Sullivan

Well, this is the philosophy on which the National Observatories operate now, of course, where the user comes in for a few days and has manuals given to him and so forth. And you were thinking then this was your policy also?

Ewen

That was my plan at the very beginning and I did not see that that much was being gained by having the students run out to Agassiz, diddle with the knobs and more than anything else upset the instrument than to just hire one engineer who would worry about keeping that system in operation at all times and assure them... come on, I would rather spend more time describing the details and the philosophy of the approach or provide assurance that is external to the system. I’d like him to perform a measurement to assure himself that the box is really performing its function than just pushing a button that says: "CAL"- I wouldn’t believe that. So the idea of we all have to run out to what I termed "Agony Station" [Sullivan: Agassiz] and diddle with knobs, I found not really the way to go. And I was very happy when Haystack came on line with computer operated control. I suggested even in the days of NEROC that perhaps what we should do or at Green Bank- I liked the idea of Green Bank and Charlottesville, I really pushed hard for that because I was on the committee for site selection for the National Radio Astronomy Observatory. And the game was we were going to find the quietest possible site. Well, there are various forces that enter into selecting a site. There were also forces that suggested that we should be in some of the best deer hunting country and very nice places like that. Why don’t we just pick one thing and focus our attention on radio astronomy like radio noise, quiet from a radio standpoint? And then consider for human beings that there would be perhaps some other place that we should locate the people who were going to operate on the data. It is very important that they have some place to get together and talk- not a place to go and diddle with knobs, when they are not that familiar with what's behind the panel. Again, I don't think that they should be that familiar with hardware. That's really asking a "mountain" to perform another "mountain" without recognizing the "valley" that exists. You would never get the "mountain" strong enough.

Sullivan

But, of course, the British and Australians were precisely those sort of people that built and interpreted the data.

Ewen

Well, a fortunate circumstance, I think. In both cases probably just a different approach in education that they were able to go down that road. I would think depending on the particular phenomena and also the, of course, as we mentioned earlier, the club type arrangement- we had the phenomenon of clubs here more strongly in place- astronomers, engineers, and so on, physicists- where you don’t see that, at least I haven’t in my visits to Australia and Europe...

Sullivan

More interdisciplinary approach...

Ewen

Absolutely. There was more mutual respect between disciplines, there isn’t the looking down on engineering. I could even feel that at Harvard. We had Cruft and we have Jefferson and of course, at Jefferson we have the great physicists and at Cruft we have these so-called engineers. And that sort of clubby type class system, it’s very, very American for some reason. I think all going away now but then it was pretty horrendous.

Sullivan

While we are touching on this, could you comment on why the US really didn't get going in radio astronomy despite the Radiation Lab and all the radar development during the war, like the British and the Australians did. What do you think was the ingredient that was missing, if that's what it was?

Ewen

I think there was as much interest in the U.S. among the disciplines that were capable of moving ahead as rapidly as England or Australia. The missing ingredient in the U.S. was the recognition of the value of this as a separate science or as a combination of important disciplines. There was no "requirement" for it. And we were, in those days right after World War II and immediately following into the Korean War and then throughout the ‘50s, searching for "requirements" to be met and it became game of a government laboratory and the Defense Department in particular to demand that a "requirement" be fulfilled by any research. So you almost had to guess the future and then demonstrate your ESP brilliance. And many proposals were filled with this type of hogwash that the Defense Department insisted on and this was all to benefit the taxpayer, to assure that...

Sullivan

It was all relevant.

Ewen

Yeah, just so that we wouldn’t get this Proxmire funding that I’m glad Proxmire finally got his funding

Sullivan

So, really what you're saying is that the Cold War, the military domination of scientific funding was the difference between this country and the others.

Ewen

That was it- focus on requirements and there was no requirement for this. And as far as the work that was done in Australia and England, it was motivated by the need for better knowledge of this new emerging field. This was very pure, very basic, very natural.

Sullivan

Right, there was no military funding.

Ewen

The thing that I found very interesting was that though there appeared to be no interest in this- lack of publications, meetings of disciplines that you would imagine would be active here, etc.- there was underground here and it was quite active. And people at NRL and AFCRL [Air Force Cambridge Research Laboratory] and within the Universities were pursuing this, but were not advertising it. When the spark- as much as the respect was allowed in the discipline, suddenly overnight the U.S. came out of the woodwork. We had all of these research facilities immediately ready to go. NRL had plans, antennas were going up and AFCRL with their solar work. So what were they doing in the mean time? They were able to pursue the research despite the fact that they were held back by these "requirements".

Sullivan

When do you see this blossoming period?

Ewen

The early 50's.

Sullivan

The early ‘50s.

Ewen

Right. I think throughout the ‘50s things took off. Once we were able to say that Harvard for example is giving a course for a doctorate in radio astronomy, that papers are being published by California, Michigan and the Government Labs- that everyone is coming together and recognizing this. That the American Astronomical Society is going to admit that engineers existed. That all of this shortly thereafter, just each little piece of it fell together quickly.

Sullivan

There was a meeting in 1954 in Washington, D.C. that was sponsored by the National Academy of Sciences, I believe, about radio astronomy in the U.S. I don't know if you actually went to it. Do you know the one I mean?

Ewen

I may have. I don't recall.

Sullivan

It was in Washington, D.C. The proceedings were published in JGR [Journal of Geophysical Research]. I was going to ask you if you thought it was an important meeting or not in the terms of the way things went. They came out with strong recommendation for a national observatory, decrying how the U.S. had lagged behind and so forth.

Ewen

I remember being on several committees with fellows like [Lester C.] Van Atta and Merle Tuve...

Sullivan

I think this was probably one of those early committees that decided to have such a meeting.

Ewen

Right and trying to get the National Observatory off the ground. I think the biggest problem at that time was the regional impact. If we are going to put up a large antenna in West Virginia, what are we going to do for California and at Harvard? That was part of the problem with the VLA.

Sullivan

Sure, that sort of thing continues between the universities and the national labs, of course.

Ewen

I think what really made radio astronomy more than anything else was Bart Bok's sheer guts to make it happen. Prolific writer...

Sullivan

Oh yes. He wrote all sorts of PR articles.

Ewen

He was going at it all the time. Just grinding it out, just insisting that it was going to happen. I recalled, well you asked about this Ewen-Knight-Dae, where I was at Harvard. He had me running around doing so many different things and ust getting the first antenna up there and pasting that together down in Cochicuate and we did that down at the D.S. Kennedy Company. It was 24 feet in diameter. We had no idea of how we were going to move it. Bart said we didn't have enough money to build a template and we can’t therefore bolt it together later. So I said, "How are we going to move it?" "That’s your job." "Alright, Bart." So I went off on a trip to Australia and when I got back Kennedy had the 24 foot antenna ready to go. So I called and said, "We’ll take it up 128." The company said, "No way, we are not going to drive up 128. It’s impossible- congestion, traffic." So we’ll fly it, helicopter, make the arrangements to fly it up with a chopper. And I recall the phone call very well. I got on the phone to the Vice President of the Harvard Corporation, "I want to talk to Doc Ewen." I said, "Speaking." He said, "Which one am I talking to? Is this the president of the Ewen-Knight Corporation, the president of the Ewen-Dae Corporation, the member of the Harvard faculty, or the guy that is going to fly an antenna with a helicopter over the top of Cambridge? Which one of these guys are you? Are they all the same?" "They are all the same." He said, "I was afraid of that." He said, "It’s all the same guy." He said, "I just want to tell the fellows in charge of the helicopter operation that if you drop that antenna on anyone’s head, you better have it insured for $2.5 million." And so that meant I didn’t get it up there with a helicopter. But it was at some point Bart said, "Just figure out how you are going to do it." So I said, "We’ll get a Harvard governor in and we’ll have it made." And we got [Christian] Herter in, went down, boom, we ran it up 128 in a truck.

Sullivan

Which governor was it again?

Ewen

Herter

Sullivan

Herter.

Ewen

Yeah, he let us take it by truck. We had to wait until the election went through

Sullivan

We got off the subject of the style of doing radio astronomy, more or less. Do you think I'm being unfair by saying that some things were over-interpreted in the data that came out of Agassiz, both with the first dish and the second? Of course, you weren’t so involved with that side of things.

Ewen

My feeling was, looking back, this is the difficult part of it- the question you ask involves an assessment after the fact, which is the way I had assessed it anyway. While it was going on, I was preoccupied with Agassiz and its operation and that sort of thing but I think a good deal of that could have been brought about by about by what might have been a natural anticipation concerning certain models. It’s not common just to that era.

Sullivan

Oh sure. You mean wanting something to be there?

Ewen

Right and that’s not in large measure just the student. Although I know in my own case, as we discussed earlier, I was very convinced that it was there in three days of the discovery, but Purcell was not, which is far more realistic. Now the reverse, I'm sure, has happened historically and I happened to have been favored with a conservative advisor. I think it would be not appropriate for me to comment on what the motivation was, only that I feel that it was not so much as an instrument problem as a motivation in the Astronomy Department. I feel that it probably was overactive anticipation. [Ewen: A perfectly normal reaction.]

Sullivan

So, you might say that there was considerable enthusiasm about running with these results?

Ewen

Because we were certainly in a competitive race at the time - once the Dutch gauntlet had been dropped.

Sullivan

Now, you mentioned that before the line was detected that you went around to AAS and URSI meetings and found them not of much use. But in the years when you and Bok were running the show there, you went to many AAS meetings it looks like. Did you find that you were then becoming accepted?

Ewen

Yes.

Sullivan

Was it Bok, the traditional astronomer, had bridged the gap?

Ewen

Yes. He kept pounding on that door, and said, "Open the door." Of course, internationally in the IAU, Oort was a "Bok" and so the doors were just blown open by those two Dutchman- Oort and Bok.

Sullivan

What about your relationship with the Dutch? I guess you were not so much in on the decisions about the science that was to be done. "Where shall we look?"

Ewen

That was Bok.

Sullivan

You really weren't part of that, ok.

Ewen

No.

Sullivan

But were you influenced at all about what was going on with the Dutch and the Australians. Or for instance, did you pick up any engineering things from Muller and the Australians, or did you pretty much operate by yourself?

Ewen

I did visit Holland...

Sullivan

Dwingeloo?

Ewen

Right and those in Australia. I didn't see anything in Holland or Australia that competitively would really make that much difference. There was so much that was going on here in electronics, particularly in the Boston area, that it became more a question of the radio astronomy community being able to afford what was available. It is pretty much the same today. So I see fellows struggling to build a 95 GHz. You can buy it for $10,000 but...

Sullivan

They just don’t have the money.

Ewen

They’ll say, "I just don’t have the $10,000." So they’ll have some student over in the corner discovering whether he can get his chip working or...

Sullivan

Well, this is the same story that society will give to the military but not to a scientist.

Ewen

And I think that from that standpoint, my familiarity with that side of it, which if I were completely an academic, if I weren’t looking to the other side or have access to that, my whole life would have been a good deal different. So rather frequently, I do get calls from people in academia that want to know what is really going on. At that time for example in ’50 [???] historically, the first maser was successfully demonstrated by Lincoln Lab. Ninety days later, and the first in industry, was demonstrated by Ewen-Knight.

Sullivan

I noticed you had an article here on masers?

Ewen

Well, that wasn’t really on masers. Well, it says masers in the title but it is more on problems you have if you have a maser. And many people have commented, not so much on the title, but it was the first paper that really took a look at all of those and we were working on the details of how to measure the noise figure of a maser, the gas discharge.

Sullivan

We are talking about a paper in Microwave Journal in 1959.

Ewen

Right.

Sullivan

So Ewen-Knight had one going at a very early stage?

Ewen

We were second on the air in eight days and Bell followed us eight months later. So now, the way that was done- and then Harvard got on the air, I guess, right about the time Bell did- so here was Nicolaas Bloembergenwho couldn’t get his maser on. Lincoln got it running. But by sitting in the outer loop, it was very simple and this was a typical example that Ewen-Knight was deeply involved in the Polaris Program at that time, ’55-’60. We were doing something on the order of $10 million worth of research in navigation, using radio starts for celestial navigation under foul weather conditions, and we were the central part of getting the azimuth right. As Red Rayburn or the Admiral used to say, "In Polaris, I know exactly where I am because if I don’t get my azimuth right, it could land on Washington." So we were then charged with the responsibility to make sure that he fired in the correct direction and under all weather and that brought on radio stars. And we were looking at radio stars at about the time NRL was publishing with Haddock and with the 50 footer. That’d seen a few radio stars. You see those- well, you can probably appreciate those in terms of X Band on the 28 footer- that’s [centroid sensing and sequential lobing?] are Cas A and Taurus. And that’s radiometric sextant at its best. That’s about as good...

Sullivan

This is beam switching so first you get positive and then negative.

Ewen

Yes. It’s just scanning constantly

Sullivan

But now the submarine came to the surface with a dish?

Ewen

No, this was brought up. The submarine stayed belong and this came up on a post and a [?] and all that good news and the transmission line was something like 75 feet long and that did not include all the bends and quick disconnects. You see if we had to submerge immediately, we had to have a quick disconnect and a guide. And this had to be very low transmission line so I transitions of X band into C band in circular guide and then gather it all back into X band, get it into the receiver that was way down in the bottom deck. So while I was building these things for $10K- $15K- well, when the maser was first announced by Bloembergen, Red Rayburn called and said, "Doc, we need low noise. If there is one thing you’ve told me about radiometry its low noise." I said, "Yes sir." I smiled and [???]. So I said, "Come on, I got to have one of these. It sounds terrific, great gadget." I said, "Let’s be second. Let’s not be first because up to this point it’s a theory." Soon as Lincoln Lab got on the air, he got on the phone, "Doc, they did it. When do I get mine?" I said, "Admiral, it’s expensive." "How much?" "$1 million" "Come on down. I’ve got $1 million." I went down and he said, "I want this guy on contract that pays $1 million and I want a maser quick." Alright and I said, "A maser is going to give you low noise amplifiers. It’s way down in the bottom of the ship. You’ve got all of this transmission line. We’d need liquid helium. You don’t have closed liquid helium systems." "What are those?" "Liquid helium, like [???]. When you’re out there in the submarine, you can’t say, ‘Hey, bring me one of those them there..."

Sullivan

...Liquid helium.

Ewen

No, he said, "Let’s build one. Let’s get on that. Let’s have a closed helium system." Then I got that going. So I had guys under contract building a closed liquid helium system and get the maser going. I called Lincoln Lab and said, "What did you use to get the maser going? What did you use?" They said, "[Potassium-cobalt cyanide?], just a simple minded crystal. You can build it in a home refrigerator." I called Nicolaas Bloembergen and said, "Why aren’t you using [potassium-cobalt cyanide?]" "Well, we don’t know how to do it. There is a trick to it." Back to Lincoln, "What’s the trick?" "A fellow by the name of Professor [Wolf?] over at BU, he knows how to do it and he’s got the refrigerator." I called [Wolf?]. I said, "You just got hired by the Ewen-Knight Corporation. Bring your refrigerator. We are building [potassium-cobalt cyanide?]." We had crystals out the kazoo and they didn’t care. So I sent him to Harvard and helped him. I am a Harvard man. We can’t say we [???], "So here you are. Put a maser on the air."

Sullivan

Now, which sources were being used by the Polaris submarines?

Ewen

The Polaris was originally a ship launched missile and it was a Jupiter missile. The navigation system therefore was ship-born and it used a 28 foot antenna and the prototype of that was the one we assembled and put in at Needham. The radio stars available to that were Cas, Taurus, Virgo...

Sullivan

Just the big half dozen there?

Ewen

Right, exactly and adequately dispersed to be good navigational aids. The next step in Polaris then was to move it to the submarine. There we didn’t have the advantage of 28 feet of aperture. So we were moved to about a 1 meter aperture and settled on the centroid of the Sun and the Moon. As a result, I became very interested in the centroid of the Sun and the centroid of the Moon and whether we were going to be seeing relationships in the centroid.

Sullivan

Or if a flare went off on one side of the Sun?

Ewen

Right and where would you look for a very quiet band? In those days in the late ‘50s, the quietest band we were able to find where we weren’t going to be interfered with atmospherically and where we could also get a very broad bandwidth through the virtue of low noise traveling wave tubes was 7.5 to 8 GHz. So we blocked that out for the U.S. Navy and immediately Lincoln Lab blocked that out for the Deep Space Communications System. So that was a very quiet place.

Sullivan

In fact, you developed along with- Frank Drake was just telling me this a few months ago- the traveling wave tube for 8 GHz, this apparently was the same receiver.

Ewen

Right, the same one was built for Polaris.

Sullivan

With a bandwidth of about 1 GHz.

Ewen

Exactly. That's the one we used to get the rings of Saturn and the comparison with the Jupiter intensity.

Sullivan

In ending can you tell me a little bit about the actual receivers that you delivered over to Agassiz, and what improvements were made during those five years or so?

Ewen

The approach taken on the receiver was primarily in packaging. Early on and as it continued and that was to make it easy for the astronomer to understand what the radiometer was all about. One of early, novel approaches which we continued throughout was to engrave the block diagram on the front panel...

Sullivan

Which NRAO still does.

Ewen

And put the knob in that part of the circuit so it is very easy to explain to anyone coming online just exactly what is happening in there. Other than that, very little was done in the way of improving noise figure. It remained the crystal mixer, the 1N21. Of course it was done in coax rather than waveguide. I would say improvements possibly for those that understood those who understood how it operated, the gain modulator was an interesting advance. A number of persons, particularly astronomers, however, did not understand some of the problems a gain modulator can carry you into encouraged me within 5 years after that was developed to drop it as a technique. Perhaps I became most convinced by the group in British Columbia who did not know what it was at all...

Sullivan

Penticton?

Ewen

Yes, Penticton and just tore the whole machine apart trying to find out what that was all about. So I said, "Never again." It was written up by a group at NRAO but we did nothing further with it. [Ewen: That technique was written up by NRAO with an introductory comment to the effect: "Look what we found under the hood of one of Ewen's radiometers." We did nothing further with it.] So until the paramp came along some years later and well after we were out of the business of providing radio telescopes. We did provide some radio telescopes with masers to the military, but not to academia.

Sullivan

Just too expensive.

Ewen

Right, absolutely. So I would say our contributions to that field ended with one technique. Knowing we couldn't improve noise figure nor could we slow the earth down, the time was pretty much set for us, we’ll go to multichannel and I think we were probably the first to attempt that and followed shortly thereafter by the group at Carnegie...

Sullivan

The Australians also were putting together a multichannel thing.

Ewen

Right. So we tried for 100 channels with bandwidths from 1 to 10 kHz and digital outputs with a driver 2 typewriter. And that was an effort we launched in '54 and it really "launched" us! As I recall, the NSF grant was for something like $40,000 to build that. We spent an awful lot of time on that. We learned a lot. The education should have probably been documented and that’s something I think all of radio astronomy would have benefitted from and probably still would, is if we could teach the engineers that are coming along that are doing the building, to document mistakes. That would be so helpful.

Sullivan

That’s true in all of life actually.

Ewen

If we could only have a publication for mistakes only with the guts to do it. So that was our big mistake.

Sullivan

What were the problems with it?

Ewen

It was just a matter of inadequate funds to do a prototype study in adequate detail to determine if there would be any problems. There was the need to go onto a schedule where you commit to a design. And if you made a mistake on paper, it’s too late. You don’t have the money to buy the... so it was just a horrendous kludge.

Sullivan

Well from a management point of view...

[Tape malfunction]

Ewen

Now from the technical standpoint, the problem was primarily the determination of the zeros baseline between the various channels and that was because we did not have contiguous filtering as a consequence of the selection of the very narrow bandwidth.

Sullivan

But the centers 5 kHz apart.

Ewen

They were possible more as I think they might have been as much as 20 or 30 kHz. Even the 10 had to be scanned.

Sullivan

Because you wanted to cover the entire Galactic line in one shot.

Ewen

Right. We were trying to get as much bandwidth as possible with a small amount of tuning fill in the gaps and hopefully to operate on a broad resonant characteristic of the frontend so the baseline would not be distorted. And that sunk. It didn’t work. We struggled with it, let me say. So it was to the point that with a lot of care and tweaking and picking the correct frequencies for the various heterodyne currents, we were able to type out a line that was realistic, correct...

Sullivan

[???]

Ewen

We did that by everyone inhaling and exhaling at the same time.

Sullivan

And your hat and your shirt hung up on that rack.

Ewen

We took it all down and said, "Forget it." We put it back on a strip chart. So other than that, I’d say it seems like low noise amplifiers and computer processing, getting rid of tail ends and doing the job right, I think that now we are getting close to a point as the speed of computers builds up and as hopefully the Josephson effect can be used by the [???]. Then we can tie a computer to the tail end of an antenna and we’ve got something. [Ewen: If I were to search for a "bottom line", I failed to perceive the difference between a grant for research in support of grad students and a grant in support of an engineering technique investigation. The availability of funds for the latter was unfamiliar to me and as a result Ewen-Knight operated on a fixed-price, fixed-time basis with academia. Harvard always wanted the new and better machine first, so the risks and blunders were borne by Harvard. Ewen-Knight, as well, was frequently bent out of shape by these moments of technological advancement.]

Sullivan

What was the noise figure of your initial 21 centimeter radiometer?

Ewen

I’d say it was around 10db, 3000° noise temperature. We probably never did see better than 2000° until the late '50's, while in our same labs we were playing with much better masers which could not be afforded by academia.

Sullivan

Something that just occurred to me is a picture of you and other 21 cm people at the URSI meeting Sydney in 1952. Can you tell me what your impressions were of the Australian group at that time? You probably toured around all the sites.

Ewen

My first impression was a very gung-ho group. Just a wonderful group of people particularly Paul Wild. It was a thrill chatting with him, interested in so many things. And of course, Pawsey was there and...

Sullivan

Bolton?

Ewen

Bolton, yeah, and that was before Bolton came here.

Sullivan

Oh yeah, and Mills?

Ewen

Mills, It was a meeting with eager minds just delighted with the opportunity to get together and chat about what we were really going in this field . A meeting that at that time was not typical of what you would find here in the U.S. in the field even at that time, a year later, or even five years later. Just an excitement. There was a camaraderie. It was wonderful.

Sullivan

What about the equipment you saw around there? Was it sort of shoestring stuff?

Ewen

It was shoestring stuff, but it was cute, a lot of cute tricks. They had thought it through. They wouldn't waste much time in the hardware areas where it wasn't all that important, to make it look pretty or whatever. But wherever a part was critical to the operation of the device, they spent a lot of time thinking about it and they had a lot of innovative techniques that they’d use. Just a feeling of someone who wants to do a good job, knows where you have to exert the effort and is very familiar with what counts and what doesn’t. It was a just very, very pleasant experience.

Sullivan

OK. That pretty well covers I think up until 1960. Thank you. That ends the interview with Harold "Doc" Ewen at his home in western Mass. on 12 August 1979.

Citation

Papers of Woodruff T. Sullivan III, “Interview with Harold "Doc" Irving Ewen,” NRAO Archives, accessed April 11, 2021, https://www.nrao.edu/archives/items/show/14880.