Interview with Marshall H. Cohen
Description
Creator
Rights
Type
Interviewer
Interviewee
Location
Original Format of Digital Item
Duration
Interview Date
Interview Topics
Notes
The interview listed below was either transcribed as part of Sullivan's research for his book, Cosmic Noise: A History of 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
Unit
Transcription
Sullivan
Ok, this is talking with Marshall Cohen in Groningen on 11 August ’78. Could you tell me what your background was and then how you first came in contact with radio astronomy?
Cohen
Sure. At first I was studying electrical engineering at Ohio State in Columbus and had a Bachelor's degree in Engineering and I switched to Physics as a graduate student at Ohio State in Columbus. And something like in a year or two after I was in graduate school, I got connected with the Antenna Laboratory at Ohio State. I worked on things connected with aircraft antennas and radar back scattering; many of these projects were done for Wright-Patterson Air Force Base.
Sullivan
Were you connected with [John D.] Kraus at all?
Cohen
No, I wasn't connected with Kraus all that time. That's all so very interesting because he had what was at the time a big array of helices right in the back door approximately. And he was regarded as the big father of antennas and things like that at Ohio State, but he had no formal connection with the Antenna Laboratory. Dick Brumsy was Director of the Laboratory. Kraus was doing other things - he had different set of students. I knew those people and vice versa, but we had very little cross connection, very little, remarkably little now that I look back on it - it was really a disjointed set of people and a disjointed piece of work between the Antenna Laboratory and Kraus' operation. My connection with radio astronomy came a couple of years later. I graduated from Ohio State with a degree in Physics, although the subjects really were classical electromagnetic theory, questions of the size of the antennas, and theoretical things like attempts to calculate the echo from things like special geometric shapes that you can think of as making up sections of airplanes or missiles and that kind of thing. And I stayed at Ohio State as what was called a Research Associate or something.
Sullivan
Post-doc?
Cohen
Post-doc of some sort after I graduated for a year. But I didn't, they wanted to me stay there and there were guys that I was in school who did stay the way I did and they've stayed there forever - they never left the Antenna Laboratory and they're sort of permanent full time research engineers I suppose you'd call them- that’s probably the best description. But life really must consist of more than just staying at one place, and I began to look for a job more or less as soon as I graduated even though I had a perfectly good job there in terms of pay and so on. But anyway, what happened is that about a year after I graduated, a letter was circulated around the Electrical Engineering Department at Ohio State. By that time I had a position of some sort in electrical engineering and this laboratory was officially part of the Electrical Engineering Department but it had students from many areas- I had done it from physics and there were others from physics and from mathematics, applied mathematics actually. But anyway, he letter was from Cornell University- standard letter- and it said that Cornell was looking for a young man, I don't what it said "great promise" or something. "We need an assistant professor and someone to work, would be interested in working in radio astronomy," because they had a project there. Charlie [Charles E.] Burrows and Joe Gordon that started up the solar radio astronomy project, they had other things they were doing also. They needed more people and they especially wanted somebody to spend quite a bit of time on it. So this looked like it might be very interesting and I applied for that job and got it. That was in 1954. So I went to Cornell in the bottom of 1954.
Sullivan
What was going on when you got there?
Cohen
Well, Burrows was chairman of the Electrical Engineering Department and he was also the father of this radio astronomy project but he spent almost no time at it, of course. Bill [William E.] Gordon was heavily involved, but he was the principal investigator or something- I don't know what words we used then, of some radio propagation experiments. There was obviously a shortage of people working on this radio astronomy. What they had done by 1954, they'd been at it for several years.
Sullivan
[Charles L.] Seeger had left.
Cohen
Charlie Seeger had left before I got there. That's probably what sparked that letter saying they needed people and they decided that instead of getting a guy like Seeger to work full time on it as an engineering type or whatever permanently than that, they wanted someone on the faculty. The project at that time had been reduced to a state that was pretty much routine data collection. They had this system at Ithaca and they were also just finishing this system in New Mexico.
Sullivan
At Sac Peak?
Cohen
At Sac Peak, it was the old Cornell gadget on Sac Peak. Let's see they had 200 megahertz receiver and there was a chain of these around the world that had been set up already five years before that I suppose.
Sullivan
By the Air Force, wasn't it?
Cohen
Well, no, let me, there was actually three or four things going on, let me see if I can remember what they all were. There was 200 megahertz receiving system already in Ithaca. There was also another one at Sac Peak. Sac Peak had several frequencies, I think. But the one in Ithaca in particular was run as a routine patrol and they got down the number of bursts per hour and the intensities and at various indices and they were sent through URSI [International Union of Radio Science] channels I think it was, and they were collected out of the [National] Bureau of Standards and bulletins of solar activity were being published regularly and I think, in fact, a lot of good science came out of those early studies of solar activity. Cornell was one of the participating stations; there was a station in Japan, in France and Australia, certainly. But even at that time, 1954, that work was being over taken and swamped by the spectroscopic study from Australia. Looking at the drifts of solar bursts in particular, those studies were going on. And that kind of routine collection in some sense still goes on in short wavelengths and...
Sullivan
Even long wavelengths. The Sun is still monitored.
Cohen
Yes, well, long wavelengths, I don't think anybody fools around with single frequency, it's all spectroscopic.
Sullivan
In terms of monitoring they do. I've just been looking at some recent reports and you can find single fluxes at 200 megahertz.
Cohen
That still goes on?
Sullivan
Yes, but now it's obscure places like Manila. People have a thing going and they...
Cohen
And they just keep going; you can always compare sunspots with indices at 200 megahertz activity.
Sullivan
But anyway, so this thing...
[Tape break]
Cohen
Chart records of the solar radiation at 8.2 cm; now the whole machinery was shipped down to Sac Peak. I made two trips to Sac Peak, I think two. And then the whole Sac Peak operation closed down the next year by the Air Force and that's when Harvard began the Fort Davis operation. Sac Peak was a terribly noisy place - it was up at a solar observatory. It was on a mountain top and you could see 500 miles in all directions. You could see to Mexico - you could see Las Cruces. It was a phenomenally bad place for what wanted to be a radio astronomy station. So then they took this thing down in the valley but they established and observatory at Fort Davis. That was very much better, of course. I had no more to do with that operation. Then the place was closed down. But then, so anyway, I was working with that. I got interested in polarization. In fact, those papers...
Sullivan
Right this one about Hatanaka’s theory.
Cohen
How did that come about- well it turned out that Hatanaka had been a good friend of Bill Gordon. I'm not sure where that originally came from, I just don't know now why they were so close. But he was a sometime visitor to Cornell, and came and spent a year once. He was very interested in polarization measurements of the Sun. He introduced me to the subject and so then I began reading Chandrasekhar’s book on polarization. I thought that was a terribly interesting subject and so we decided at Cornell to build a radio polarimeter, but not a kind that simply was a rotating dish but was a correlation-type polarimeter. He built that and it was a student of Hatanaka or someone who was from Tokyo, a guy named Kenji Akabane...
Sullivan
Akabane, yes.
Cohen
Who came over and stayed eighteen months. He then went on to Michigan for six months or so and worked with Fred Haddock, must have been 1961 or something like that. I don't remember the years now, but it could be reconstructed by looking at the papers. But Kenji and I studied this business; I had actually built up some aspects of the polarimeter prior to Kenji's coming over, but then we worked on it. I know now in retrospect that we did a couple of things quite wrong. We had very bad signal to noise characteristics in correlation channels because we didn't do the right thing. But nonetheless, I think it was a good attempt to try to do correlation polarimetry in 1958, which is when that was.
Sullivan
Let me ask you about this polarimeter, whether there were particular difficulties in building it or was it just a new idea to try this business with too narrow bandwidths and so forth?
Cohen
The idea, well, first, along with Hatanaka, we'd been doing a lot of calculating of transfer theory, the characteristics of the radiation- a burst generated near the plasma level deep in the sun. Radiation has come up out of the Sun - what happens to it? Suppose you start with a linear polarized signal so then you get a Faraday dispersion and if you look at a wide bandwidth you're going to get it all smeared up and won't see the polarization. Because Hatanaka, I believe, who first speculated on the existence of this polarization smearing within the bandwidth on a ray coming up out of the Sun because of the possibility it was some enormous magnetized plasma and we did a lot of model calculations and you can calculate thousands of Faraday rotations in some cases. So we tried to think of ways of trying to measure this, and one way, obviously, is to instead of measuring over a broad band, you measure over a narrow band and see what happens. So we just picked two narrow bands. It just seemed like an obvious thing to do once we'd thought about how the rotation might exist. Of course, narrow bandwidth makes other problems. Those, I wrote in 1958, the IRE [Institute of Radio Engineers] Proceedings on Radio [?]. I think Fred Haddock was the editor of that. I wrote two papers for that. I've been surprised over the years, those papers get referred to. I see references to those.
Sullivan
Oh yeah, the one on radio astronomy polarization measurements is a classic, if I may say so.
Cohen
That one gets referred to - I keep seeing it, references to it.
Sullivan
It was the first time that it was all laid out.
Cohen
I tried at that time, the thing that I, that struck me at that time was the number of independent quantities that there were and why therefore, there's four numbers to specify the fields. And so you need four measurements. But they have to be independent, and upon trying to read what the literature said, Kraus' books on polarization and so - most of which dealt not with the random field but with a sinusoidal field. It was clear that a lot of what was being said about polarization involved things that you measure which were not independent. And so you do all these things and you end up and can't calculate what you want because something is the sum of something else, whatever. And so that was what led me into drawing page after page of pictures. Whole desks full of little pictures with arrows, and things rotating and what adds up to give what and so on. It was all these geometric reconstructions of what it is that's independent that, and that article is written that way.
Sullivan
This is the outgrowth of that.
Cohen
The outgrowth of that. There were just hundreds of little pictures like in that article, of course, but what came out of it was an attempt to classify the methods of making polarization measurements on the numbers and quantities that we needed to measure and ones that you can measure. If you have a priori knowledge you know there's no circular polarization, what's then the best way to make a measurement, you want to measure weak rotation or the other way around. So I tried to classify in my own mind all that stuff or as much of it as I could; that's what I wrote into that paper.
Sullivan
Why do you think that there had not been that much polarimetry done on solar work before? Of course, there had been some, but never really detailed studies.
Cohen
There had been some. I don't know. It was about that time that there was an enormous growth of interest in plasma physics in the 1950s, and all the business about Faraday rotation and the different kinds of waves was a very important topic in the 1950s. That must have penetrated everybody's consciousness in some way or another and led to the notion of polarization.
Sullivan
I'm going to go back one step further, why was there this growth in plasma physics interest? Was this an outgrowth of military funding or something?
Cohen
Probably. It was about that time also that I got friendly with people in Paris. I met the people from the Paris Observatory first at the Boulder URSI meeting whenever that was.
Sullivan
1958, I think?
Cohen
Must have been before then.
Sullivan
1957 maybe? Anyway...
Cohen
That can be reconstructed and I was struck that they were talking a lot about plasma physics and [Jean-François] Denisse...
Sullivan
Which people do you mean? Denisse and [Jean-Louis] Steinberg.
Cohen
And Steinberg and [Emile-Jacques] Blum. And I've been friendly with all those people. Since Blum, is here today, I did a lot of talking with him already. And in fact, we've remained friendly with a number of those people for about twenty years. But I got a lot of interest in plasma physics from talking with Hatanaka. He had the notion of polarization being important in studying solar bursts. I think maybe before anyone else, I don't know. And there was the business of trying to follow a ray as it comes out from the center from some point in the Sun and there's a dipole field above it - what happens to the rays- its the question of polarization that comes out. But to study that, there are all kinds of other things that come into it. And it turns out that the same problem existed in the ionosphere or many of them. Cornell was a very interesting place for these kinds of studies because Henry Booker was there and Bill Gordon and that big ionosphere group. So they were concerned with propagation through the ionosphere and they were doing all of these elaborate kinds of things of study and rays reflected from the ionosphere, triple splitting and a lot of other interesting and bizarre phenomena connected with magneto-ionic theory. And Henry, Henry Booker regularly gave a set of brilliant, brilliant isn't the word, beautiful lectures on magneto-ionic theory. He's one of the best lecturers in the world, Henry Booker is, things are just absolutely crystal clear when he's talking. I suppose that got me interested to an awful lot of these aspects of magneto-ionic theory, and it was about that time, also, that some of these other questions connected with these rays that come up out of the Sun came into being. I was trying to calculate what polarization comes out of the Sun, it's the same question as the polarization that comes down from the ionosphere, and you get coupling. If there's gradients in electron density or magnetic field or in well, I guess those are new parameters, gradients in the index of refraction. What it means is that you have to be very careful. Sometimes the modes don't propagate independently, you don't always get Faraday rotations and these other things. And you can ask some very subtle questions about what the state of polarization - where does the polarization get locked, do you still have Faraday rotation or is the polarization still changing slowly when you're out nearly empty space. The answer is no, for various reasons. But anyway, where does the polarization get locked, where does the polarization get locked when you're coming up out of the Sun? Or does it go through the Earth's atmosphere? Why is it polarization doesn't change, suppose you have a particular club of polarization that just happens to match one of the mode in the Earth's ionosphere, but then as you propagate through the magnetic field, there's change, you get some twists in the Earth's ionosphere. Maybe what we measure isn't what the Sun put out, but it's an artifact to the ionosphere. Turns out that's almost certainly not the case, frequencies when you look at it are too high. Anyway, there were some questions like that, and I talked with Booker some about that and he put me on to some of these other papers which I began reading and that was the outgrowth of an article which I was very proud of at the time though it doesn't seem to have made very much stir.
Sullivan
Which one is that?
Cohen
A paper in the Astrophysical Journal on...
Sullivan
1959 on type 3 and 5 bursts?
Cohen
No, this wasn't...
Sullivan
Oh, 1960 on magneto-ionic mode couplings.
Cohen
That it. Mode couplings. I spent a lot of time over several years on it. Theoretical paper on propagation theory, a question of when the characteristic modes are independent and what it really turned out that a great deal of it had been done before by the ionosphere group. What they hadn't done though, was allow for the magnetic field to be varied. They had been interested in variable [?].
Sullivan
Well, and also, perhaps the stronger solar field gave you different effects or was that...
Cohen
That's another matter. The only cases that I was able to solve with any or get any handle on, was the weak field case. Strong field case, actually there's been work more recently on those other cases. That's another case, there's an occasional theoretical paper still published on that kind of thing, and that paper also was referenced once every great while, I find.
Sullivan
Did you feel that a lack of basic data on the corona in terms of its magnetic field strength and electron density for you to put into any theories that you might come up with? Or did you feel like that problem was pretty well known.
Cohen
Well, that was behind a lot of this. That seemed to motivate the solar physicist. They wanted to get better estimates of coronal densities. I guess the biggest problem at that time was the temperature profile. Where did the temperature really change? It was cold down below and hot up above - where did it really change? And how uniform was it? Once you got into the corona, was it always a billion degrees or 1.3 billion or whatever the number might be. How uniform was it? I think that was the thing of interest. And the variations in the solar cycle also was very interesting. I have a little trouble at the moment in remembering what was the motivation for those solar studies. I think to some real extent a lot of that business of radio astronomers studying the Sun came about because that was the equipment available at the end of the war.
Sullivan
That was the strongest source.
Cohen
The strongest source and the equipment could do it. They had seen the Sun in 1943 or whenever it was. The people who were in on that regarded that as very interesting and an unusual thing. And people like Martin Ryle and others jumped immediately into the Sun. They never left that not long after...
Sullivan
When you could begin to do some other things.
Cohen
Began to then look at galaxies...
Sullivan
Of course the Sun continued to maintain much more practical importance than the rest of it. And there is always military funding which was...
End of Tape 112A
Sullivan Tape 112B
Sullivan
Continuing with Marshall Cohen on 11 August ’78. One thing that occurred to me is did you ever think about using any polarimetry techniques to non-solar effect and if so, why didn't you do it?
Cohen
I did to a small extent, but I never pushed the ideas very far. We had no large antenna, there was no real opportunity. Other people were doing those things anyway.
Sullivan
Now, about the Arecibo antenna, can you tell me when you first heard about it and then what role radio astronomy played in the development?
Cohen
I was very close with Bill Gordon; we talked a great deal about the Sun, about propagation sometimes. And he got the idea of using Thompson back-scattering as a means of measuring electron density, and calculating the size of the telescope, the size of the reflector that would be needed to get echoes using state of the art electronics. I think he gave a talk about it, either a formal lecture or else just a lunch conversation, as it were, on the very large telescope he was talking about, a very large dish. It was immediate then- I mean the immediate idea in my mind was that this could be extremely powerful radio telescope and could be used directly for planetary radar measurement. I made the first estimates of the return that you could get from planets. This was all within the first few weeks it seems to me.
Sullivan
But now hold it. He was thinking of a fixed dish.
Cohen
He was thinking of a fixed dish, that's right. But I, nonetheless, given the dish- well, what I did was try to calculate what you could see- if you had this dish and this transmitter and this receiver, you have the system that he was postulating that could see electrons in the ionosphere, could it see the planets? And the answer is, "Yes, it could see the planets." But of course, only if the planet was in the beam which meant that it had to be, it was a fixed dish, the thing had to be in the tropics and it could see the planets for five minutes every ten years or whatever it turned out to be. So the idea of a steerable dish came about, was introduced very early on in the history of the project because of the potential interest in looking at the planets.
Sullivan
I see. That was the primary motivation to make it steerable.
Cohen
That's right. That's how the notion of being steerable got into it and that was a very early notion. I think it was certainly within the first weeks of the original conception of the idea. The notion of using it for radio astronomy itself as a passive reflector was also there in a sense from the beginning, but not in terms of such a specific applications. That wasn't anywhere near as strongly motivated force. John Cox and I spent quite a bit of time talking about the uses of dish, this postulated dish, scientific applications of this thing, and John and I had a piece in it, and we discussed also the radar possibilities.
Sullivan
What did it look like would be most promising in the passive? Of course, I can check the report.
Cohen
I'm trying to remember. I can't remember at the moment. I remember we tried to think a lot about stars. The notion of looking at flare stars was in the air at that time.
Sullivan
Lovell had just, in fact, discovered a couple.
Cohen
I think that's right.
Sullivan
I think that was ‘57 or '58.
Cohen
And John Cox and I talked a lot about different stars and what would happen. In fact, there'd been some use of Arecibo for flare stars but not a heck of a lot. That was not a big application it turned out, of course, other things that we didn't know about have become a big application. Pulsars, for example. Had we had the wit to think about pulsars, we would have thought immediately about that.
Sullivan
You had an upper frequency now of 600 megahertz?
Cohen
The first frequency was 430 megahertz. That was probably the first one that we thought about. Well, it was a radar- it was an assigned frequency, the kind of frequency you could get for powerful space radar. And we talked about that - I think 600 megahertz was one of the early radio astronomy frequencies. But that was not built into the design of, the design of the reflector was for 400 some megahertz and it worked pretty well at 600. It did not work at 21 cm. We tried very early on to, we had a primitive 21 cm receiver and the reflector was no good.
Sullivan
But the specs for the reflector didn't say that.
Cohen
No, that was no surprise. But we recognized the importance of the system since we were studying hydrogen, but it was clearly not suitable for that, of course, it was 15 years later or whatever before the thing was resurfaced and then it began to become extremely powerful for hydrogen.
Sullivan
Last week Bill Gordon told me about the getting it funded and this sort of thing, he covered that pretty well. I was wondering from your point of view, were there necessary compromises between the passive applications and the active ones or were you quite happy with the way, from a radio astronomy point of view, that the specifications went?
Cohen
I'm trying to remember. I didn't get involved in the detailed design very much, but I was certainly involved in discussions of all kinds. There was one unfortunate thing, I mean nothing that had been designed in but it did happen, of course, that the first feed that was built was bad. It took us a lot of that first year just to find out what was wrong with it.
Sullivan
What was wrong with it?
Cohen
There were exciting higher order harmonics. It was a square feed, perhaps you know what it was it was slotted and square and 96 feet long, and the basic problem ended up being that the elementary little aperture pieces were not on the focus but were substantial fractions wavelength focuses, they had to be in that excited circumferential modes and so there were whole great areas of the aperture ultimately that were out of phase. Ron [Ronald N.] Bracewell who figured that out. In fact, there was this famous Bracewell committee that lasted for years and years.
Sullivan
I'm not sure I know about that committee.
Cohen
The Bracewell committee was a committee laid on, I suppose, by the Air Force as a watchdog committee. It wasn't really a visiting committing in the NRAO [the National Radio Astronomy Observatory] sense. It was a committee- you better ask Bill Gordon or ask Bracewell, but from the beginning there was the Bracewell committee and it had a number of people, membership circulated, but Ron Bracewell was the boss.
Sullivan
Well, it was a [?].
Cohen
Even while it was- well, the committee certainly existed at the very beginning. I don't know whether it existed at the early part of the construction or not, but it was an advisory committee. I don't know now whether it was a committee organized by the Air Force who was paying for it, or whether it was organized by Cornell University.
Sullivan
But this was one of their first major problems anyway.
Cohen
Well, Bracewell himself, spent some time in Puerto Rico. I lived there for almost a year and the thing obviously wasn't working, at the very beginning of the thing when it first came on we discovered lots and lots of double sources which turned out to be side lobes. We got to recognize what the beam shape really was and had these enormous side lobes and nobody understood the thing and then it was only after we went through some horrendous exercise of basically measuring the primary distance- the illumination right on the mesh. We broke up a little sled and wagon and so on, big experiment that I was essentially in charge of, and we had recorders and we transmitted from above and ran - might have been the other way it might have a transmitter in this little sled and we slid it up and down and made lots of radio cuts across the mesh and recorded the signal and we had phase reference. Basically, we mapped the amplitude and phase over this mesh and did Fourier transforms and found out what the aperture distribution was. We recognized that there were these sections, I think it was a six sided symmetry, pluses and minuses, it was whole great sections that didn't look right and we stared at this a long while. It was Ron Bracewell, in fact, who did most of the detailed analysis of it. You could analyze that into appropriate cylindrical component and carry these back up to the feed and then it was recognized, it sort of almost became obvious long after the fact. That feed had been designed by Allen Ked at TRG and he was no slouch. He was very highly respected antenna designer who knew a great deal and I guess that he used to be faulted ultimately for having had designed the feed which didn't work by three or four dB, that is the gain was off by three or four dB. because a third of the aperture was out of phase- that's approximately the way to look at it.
Sullivan
But of course...
Cohen
He didn't know it in advance.
Sullivan
But of course, no line feed of this size had ever been built before.
Cohen
No nothing like that. There have been some line feeds built, but they were thin line feeds. And this gets back to your original question, had this been done only for radio astronomy we wouldn't have tried to build the thing that was circularly polarized which was needed for electron scatter. We would have built a thin one and then the dipole, the effective dipoles, would have been very close to the axis. In fact, as soon as Bracewell made this breakthrough, we then began to think of what you could do to cure it, and we, very soon then, built some thin feeds. The first one that built was 600 megahertz. That worked very well.
Sullivan
I see. This would be what year now?
Cohen
Well, I lived in Puerto Rico in 1962, most of 1962 and we built a traveling wave 600 megahertz toward the end of that year. In fact, I built one with a student named Giovanni Perona. We wrote a paper on that in the IEEE [Institute of Electrical and Electronics Engineers] something or other, I suppose. That worked pretty well. Since then they've built line feeds; that was a short thing; that was a thing with dipoles and it worked well, reasonably well.
Sullivan
Well, anyway...
Cohen
By that time we were beginning to understand the feed problem. Since then, much more elaborate feeds have been built.
Sullivan
This thing was also very bad for the ionosphere mainly because of the loss of gain. Not so much the side lobes.
Cohen
That's right.
Sullivan
Although you probably didn't like those either.
Cohen
Yes, well, it was murder for the ionosphere or for radar, any radar because it's a round trip situation. We were off 60 meters or whatever it is on the round trip.
Sullivan
What in fact were the first decent, so to speak, radio passive observations that were done?
Cohen
I suppose the first serious work that was done there in my mind was interplanetary scintillations. The beam had such a bad shape, excuse me, are they running in there now? I'm supposed to be taking serious notes.
Sullivan
No, five minutes.
Cohen
Okay. The beam was so bad that you couldn't do and the change of gain with zenith angle was also very bad because it had a lot of spillover, a [?]. You couldn’t make the proper flux measurements and they were just very difficult to keep things...
Sullivan
Or it took a while to calibrated out.
Cohen
Calibrating...
Sullivan
[?]
Cohen
I guess they have that to some large extent still. The higher frequency ones at least illuminate a smaller section.
Sullivan
That's true. It's not as severe.
Cohen
It's not so severe for the first 10° or 15° of motion, but anyway, those problems were just extremely severe and the things that did work out well were those things which depended on rapid variations. We did some lunar occultations which were good. In fact, had...
Sullivan
Hazard was...
Cohen
Yes, well, actually the very first ones of those I did with [Mukul Ranjan] Kundu, even I think wrote a paper about that. [Cyril] Hazard came the following year and did a whole string of these, Hazard and [?] and others. But that was started before Hazard got there but then he built up a very much larger program. He even had, well, we got [Sebastian] von Hoerner’s algorithm on our radio and modified it some and used it at Arecibo. Then Cyril came from Australia and did his own trick. But then we got into interplanetary scintillations, wrote a whole series of papers.
Sullivan
That's getting a little bit past my... I wanted to ask though, once the feed problem got fixed, what fraction of the dish time went to passive radio astronomy?
Cohen
I think about half, I should say.
Sullivan
That much? Even that early?
Cohen
Yes. But radio astronomy always, from the beginning, took more than its share, as it were, of time. The radar guy would take very small amounts of time, that is, planetary radar was an occasional very intense operation but it lasted for only six hours or maybe they'd do practice and have to fire off the transmitter and all that, but it was basically a very short term, very intensive operation and a lot of computing. The ionosphere occasionally, they would take large blocks of time, sometimes they'd do continuous 48 hour experiments, transmitting continuously and so on. And then all the rest of us filled in in various ways with radio astronomy. A lot of the stuff in that first couple of years turned out to be no good simply because the side lobes were so bad. [?] from Paris was there for a couple of years. He got very little science out of what he did all that time. He was looking at galaxies and lots of other things, and it turned out that a great deal of that simply was of very marginal accuracy.
Sullivan
One final question. You started off essentially as an electrical engineer or half that and half physicist in the early ‘50s and then you became what we call a radio astronomer. What did you say that you did at a cocktail party say in 1957 if someone asked you?
Cohen
In 1957?
Sullivan
When you were at Cornell.
Cohen
Well, 1957, I had been in the Electrical Engineering Department for three years, I was a junior member of the EE faculty; I was in electrical engineering, but I was doing radio astronomy.
Sullivan
So you would say that you were a radio astronomer then?
Cohen
Well, when I went to a meeting of the AAS [American Astronomical Society] or URSI or something, I'd say I was a radio astronomer. I don't remember what I said at Cornell. During that period, I stopped calling myself an electrical engineer and began calling myself an astronomer, because in 1963 or something like that, I forget when, I left electrical engineering totally and went to the Astronomy Department.
Sullivan
At Cornell?
Cohen
At Cornell. I was an Associate Professor of Electrical Engineering and then I became an Associate Professor of Astronomy - just switched departments. Then it was easy...
Sullivan
No ambiguity.
Cohen
That's right. That was an interesting switch which isn't often done, but I- engineer isn't the right word but my interests just got that way and I talked to all the people, all the different chairmen and the deans and so on and we just went ahead. Wrote something down on a piece of paper and I changed my office and my teaching changed. Well, it didn't even change that much necessarily. That was, what did, the teaching loads were less in astronomy than in engineering I think that's true at every university. So that was one thing, but that wasn't what was so fundamental though I suppose some people thought it was, but what was more fundamental was the associations. My interests were really in astronomy, radio astronomy and the equipment, a lot of the equipment and techniques but also in astrophysics. The people that I really wanted to talk to and eat lunch with and so on were always in some building that was a quarter of a mile away.
Sullivan
Up till that time were you still the only radio astronomer at Cornell?
Cohen
That's almost true. When Arecibo began getting substantial there were people who were working in Puerto Rico, of course. Well, Arecibo brought in a lot more people. The solar work prior to the Arecibo time, there was me and there was, it was [Ed Shipmakker?] who left somewhere in there and went out to Bureau of Standards. There was an occasional person like Kenji Akabane.
Sullivan
But basically, though, you were the only one.
Cohen
Basically, I was the only staff member who was interested in doing any real radio astronomy.
Sullivan
Okay, thank you very much. That ends the interview with Marshall Cohen on 11 August ’78.