Interview with Kenneth I. Kellermann (with Harry van der Laan)
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
And now I’m talking to Ken Kellermann and Harry Van der Laan at NRAO also on 19th March in a tripartite discussion. Let’s first of all start with Ken. You started as a grad student at Caltech. What year?
Kellermann
’59.
Sullivan
And what was the situation at that time there?
Kellermann
The first 90 foot antenna had just been finished and I got there in the fall of ’59 during... well, I guess that had been finished some months earlier and they’d already done some single dish observations. [Daniel E.] Harris and [Morton S.] Roberts work was in progress and the [John G.] Bolton and [Robert W.] Wilson work on the galactic survey was in progress. And then during the first six months or so I was there, the interferometer got put together and they started the interferometer work. I wasn’t doing very much then.
Sullivan
What were the first interferometric observations?
Kellermann
The things were built to measure accurate positions essentially. Up until that time, I think the highest frequency interferometry had been some hundreds of megahertz, Australian stuff I guess, and Bolton and [Gordon J.] Stanley stuck their necks out to try to design something to work at 50 cm and the dishes were only sort [planned?] to work there and a lot of people thought that was way beyond what could be done.
Sullivan
This was to go for better positions?
Kellermann
Right. For identifications and in the course of building it, they decided to go from 50 cm to 30 cm and the first system that was actually working for a year and a half or two years, I guess, was at 30 cm, 960 MHz and the big project then was the positions.
Sullivan
Was it primarily Bolton’s idea that identifications was the important thing to do?
Kellermann
Yeah, he organized the whole project and that’s what he wanted it essentially to do, to do identifications and then sort of after that the [Alan T.] Moffet Matlby work on the source structure started.
Sullivan
And what was the first thing that you started working on?
Kellermann
I started on the spectra. I was just using the interferometer to measure fluxes. I was just using the closest spacing you can get but it was a very useful way to measure fluxes because, I mean, you know it was all reduced with chart recorders, you just had a series of fringes, and you average them up rather than taking [drift?] scans and having all the problems with baseline troubles. I think it was probably the first time an interferometer was used to measures fluxes and that’s why it was considerably more accurate than anything that had been done before.
Sullivan
And so your thesis was putting together the spectra of these sources?
Kellermann
Yeah, and I was very naïve, I guess, when I started. That kind of thing was working only at 30 cm. 20 cm was in the works and there were some hopes of doing some 10 cm work and I thought that was enough to study spectra.
Van der Laan
When was this Ken, in 1960?
Kellermann
Yeah, essentially. There was just a simple thing. If you had two frequencies a factor of two apart and you measured each one with an accuracy of 10%, well, you had your spectrum [?] ±1.5 or something like that. And then it was one of these real coincidences that [???], the Berkeley IAU meeting, I think...
Van der Laan
It was in 1960.
Sullivan
61’.
Kellermann
’61. [Francis] Graham Smith was there.
End of Tape 35A
Sullivan Tape 35B
Sullivan
Ok, this is continuing with Ken Kellermann and Harry Van der Laan.
Kellermann
You need a wider range of frequencies than a factor of 2. And Cambridge had been doing all these low frequency surveys. So I asked Graham about the availability of low frequency data and it turns out that some hours before that Robin Conway had been doing spectra work at 220 and 400 MHz or something at Jodrell Bank and also asked him about this. And in the course of conversation it was thought that if we could put all this stuff together, it would very useful. And well, it was Bob Long who was...
Van der Laan
That’s right.
Kellermann
I mean Bob Long also was working on the Cambridge low frequency spectral data at the time.
Van der Laan
What was it, 38 and 81 Mhz?
Kellermann
Yeah. I mean he took the survey data to work on the spectra so for about a year, I guess, we corresponded on things and sent data back and forth and everything. And then I went over there in the spring of ’62.
Sullivan
So this is while you’re still working on your thesis?
Kellermann
Yeah, I spent a few months in Cambridge where I actually put it together.
Sullivan
Still before you had your degree?
Kellermann
Yeah. That’s where I met Harry.
Sullivan
Ok, so that’s when you got to Cambridge so let’s go back to you. You were in Cambridge earlier?
Van der Laan
I went to Cambridge in 1960. I think it’s amusing, not historically, but maybe if you need to know how I got into radio astronomy because I did grammar school in Holland and then my parents immigrated when I was 16. I went to [?], of course with my whole family and I worked for a couple years, earned English at night and so forth and went to high school in grade 13 in Ontario as a fifth high school year. I went to university to do math and physics and graduated with honors, got a gold medal in [?] and got one of these things, a Woodrow Wilson scholarship to go to Stanford to do theoretical physics. But Diane and I said, "It would be sort of nice to live in Europe for awhile," and I said, "Ok, I’ll apply for some fellowship or scholarship in Britain." So I applied for a Commonwealth Fellowship which was just for the first year. It was a Commonwealth program just established and I remember how laborious it was to fill out those forms. It was in quadruplicate. I almost said, "Oh, forget it." I went through with it, getting the references, and writing sort of tell me what you might do and I had no idea. So I just applied for fellowship in mathematical physics in Cambridge and I had no idea what it might entail. So low and behold, we had already rented an apartment in Palo Alto and I got this thing to go to England and they said, "You can go and do mathematical physics at Cambridge and you will be a member Pembroke College, a research student at Pembroke College." So I arrived in Cambridge and I had no idea what to do. So for 6 weeks I shopped around and I really met some people with impressive names, Professor Frisch in nuclear physics and [?] in high energy physics, and [G. R. Taylor?], the grand old man of fluid mechanics, and [?] and [?] in [DAMTP?].
Sullivan
So you had to get one of these people to agree to take you on still?
Van der Laan
Yes and, in fact, it turned out there must have been a shortage of research students or something but I could have worked with any one of those groups as a research student apparently with the qualifications I had. With the fellowship, I didn’t need any money. I just needed an interesting project and so I went to a lot of lectures and I talked to other research students who were there. I talked to each one of these and I made appointments and Ryle was my last appointment. No, I hadn’t had [G. R. Taylor?], he wasn’t there. I’d had all the others and talked to them personally as sort of an A student. And had an appointment with Ryle [Martin Ryle] and one day I walked into Ryle’s office, into his secretary’s office, they have connecting doors- Molly, she’s still the secretary. I said, "I have an appointment with Dr. Ryle," and Molly looked and me and said, "Professor Ryle, you mean."
Kellermann
No, Mr. Ryle. He wasn’t a professor then, was he?
Van der Laan
No, he was. In ’60, he was.
Kellermann
Really?
Van der Laan
Yes, he was Professor Ryle. He wasn’t a doctor though. And of course, I never knew it made any difference. But anyway, he doesn’t have a Ph.D. of course. None of those people have. Tommy [Thomas] Gold [never got a Ph.D. either?]. I think it was fashionable in Cambridge as soon as you got a college fellowship to withdraw your candidacy from the Ph.D.
Kellermann
Well, the war also affected those people. By the time the war was over [???].
Van der Laan
PhDs were considered inferior to a fellowship at Trinity or at [King/Kings?] or something like that. If you have a global outlook and you need a Ph.D. as a bread and butter ticket but in those days it was not important. Anyway, I walk into Ryle’s office and he sort of looks me over. He didn’t have any of my records nor transcripts. He said, "Ok, go on, tell me about it." I had honestly never even heard of Ryle or radio astronomy in Cambridge. I was completely naïve. I’d always done mathematics and physics as an undergraduate. I didn’t have the vaguest notion. So Ryle said, "Ok, so you are looking for something to do?" "Yes." "How long will your fellowship last?" I said, "At least three years." He said, "Ok. Well, we’ve got this problem." He said, "Sit down," and he had a blackboard in his office. He still has the same office. Well, they’ve now moved and he started drawing supernovae remnants. I had heard about supernovas. I had never heard about a supernova remnant and he said, "These things emit radio emission." He said, "We don’t know why and we don’t know how. We want to know about it." And he said, "If you are interested in mathematical physics or theoretical physics, this is a good theoretical problem." And he said, "The last room at the end of the corridor has an empty desk so why don’t you go ahead and start." And I said, "Well, you tell me a little more about these things." So he did sort of in a vague way but I thought his enthusiasm was so contagious. And with all the other people I talked to, they were interested and interesting but they didn’t convey something that was engaging or enthusiastic. I mean, Ryle was just enthusiastic in a childlike way. He was exuberant about the stuff and that really took [me over?]. And I went home that night and said to Diane, "I’d met this interesting man, Professor Ryle. He is talking about crazy problems. He said, "I could start right away," and I don’t know what it is. I don’t know whether I could ever do it but it sounds fascinating and I am going to do it," and I did. And I remember I spent the first two months reading the Paris Symposium in Radio Astronomy which was the only sort of overview. It was published in 1958 and I started to read a few other astrophysical monthly IAU Symposium reviews. And I started to go to the lectures by [Leon ?] called cosmical gas dynamics, which was really a very good way to start. And that’s how I go into radio astronomy. At that time, I was the only student who was doing "theoretical" work in the whole group. Peter Scheuer was a theoretician but he was in Australia. For two and a half year, we only overlapped for a few months, Peter Scheuer and I. We overlapped enough that he was one of my Ph.D. - what’s it called- examiners, he and [?].
Sullivan
So your Ph.D. thesis was on supernova remnants theory?
Van der Laan
Only the first half. I mean, I, in ’62, published two papers on supernova remnants that came out of that. Then I switched over to radio galaxies, extragalactic source theory. I went to the [?] Texas Symposium. Were you there Ken?
Kellermann
No.
Van der Laan
No, Conway reported on this. He gave that paper. That was the symposium called [The Physics of Non-thermal Radio Sources?]. That preceded by about two months, the recognition of the 3C48 spectrum as being highly red shifted.
Sullivan
And when was that now?
Van der Laan
That was December, just before Christmas, 1962.
Sullivan
And where?
Van der Laan
In New York at [?] Institute in Riverside Drive.
Sullivan
Are there any proceedings from this conference?
Van der Laan
Yes. There is. There is.
Kellermann
It’s called the [Physics of Non-thermal Radio Sources?].
Sullivan
Oh, that’s what that is. [???].
Van der Laan
[???] and Greenstein was there and Martin [Sweet?] and they had very peculiar papers explain what are now quasar spectra in terms of white dwarfs.
Kellermann
There is a preprint of this paper floating around. You got to try to get a hold of that.
Van der Laan
That’s not published. They withdrew the paper.
Kellermann
That’s right. I’ve got it somewhere. I don’t know if I could find it again. All this lines at 3C48 which they identify with all these exotic [?] ionized things and what not and just as they were about to send it in, there was a breakthrough.
Sullivan
Yeah, if you should find it, send me a copy please.
Van der Laan
I had written [?] before I went to the conference and said, "Look, what I need now..."
Kellermann
[I think they burnt all the?] copies that I know of.
Van der Laan
I had a kind of theory, you know, which we knew then from Moffet and Maltby work that radio sources were doubled, tended to be doubled. We had no idea how confined they were and I was trying to develop a theory of having spherical explosion where the duplicity came in as a kind of [?] brightening due to expansion into cylindrical symmetric intergalactic magnetic field. And I worked that out and I published that in two papers [and in monthly notices] in ’63. Those four papers constituted my thesis plus the last chapter which deals with radio sources in clusters in 1963, which I didn’t publish at the time.
Sullivan
Let me ask you both a question about those days at Cambridge. That’s when log N-log S was just getting resolved, it seems to me.
Kellermann
I don’t know whether it’s been resolved yet.
Sullivan
Well, at least the Mills-Ryle [let’s say?] discrepancy. The Parkes antenna hadn’t come along quite yet and that was perhaps the final thing that did it. But what was the situation, I mean, as you saw it? Was it really sort of a fort, you know, where you are constantly firing salvos or was everyone just concerned to work it out and they didn’t care?
Kellermann
That was actually, I think, at a minimum at that time. See it reached a peak around 1960 between the Cambridge and Australian stuff.
Van der Laan
By the time I got there pretty well that battle had been fought with the Australians.
Kellermann
Yeah, that’s right. And then it didn’t really come back up again until Parkes got going.
Van der Laan
Well, except that there was, I mean, a major announcement with a lot of press coverage and a Royal Astronomical Society [RAS] meeting [???] where Ryle announced the result. These must have been the results using the 4C antenna.
Kellermann
That’s right. That’s right.
Van der Laan
And from that resulted the Ryle and Clark paper, Rupert Clark.
Sullivan
This was just the first part of the 4C which confirmed the 3C [P of V?], something like this if I remember correctly?
Van der Laan
Yes, I think you are right. I think it was all alright in the 3C [P of V?] but extend things, yes, quite a lot deeper than 3C. And of course, I remember that first of all, there was an RAS meeting, there was a press conference and an RAS meeting and I remember, Hoyle and Ryle immediately started trying to contrive the steady state cosmology with locally [quasi- Hubble?] expanding regions so that you sort of save the steady state.
Sullivan
So this was sort of a local minimum in the controversy when you were there?
Van der Laan
In ’61, ’62 it was pretty quiet. I thought it was in ’62. It may have been ’63 that this announcement was, the Clark Ryle Paper which I must admit stirred it up quite a bit. I always had the feeling that, I think in that sense, steady state made a big contribution because it was a great stimulus. It was not so much, if you like, a cosmological theory in the sense of a full-fledged theory that was being tested but it was really a test simply of the perfect cosmological principle.
Sullivan
Right. Well, it made a definite prediction which none of the others did.
Van der Laan
Well, but you don’t even need the theory for that. It’s just the hypothesis of the perfect cosmological theory.
Sullivan
Yes, that’s true. It’s unique.
Van der Laan
Forget everything about geometry.
Sullivan
And that’s what made it so vulnerable, of course, also [???].
Van der Laan
It wasn’t a terribly subtle effect. It was a terribly clear cut. And I remember how pleased Ryle was when he got [Jan Hendrik] Oort’s preprint. Well, he had many mixed feelings. Oort gave, at the symposium in Paris called [???]. Oort there called that effect the Ryle Effect and Oort was one of the first really prestigious astronomers who completely accepted his first order interpretations of his radio source counts, called it the Ryle Effect, and worked it out. And I remember he said, because I walked into his office when he had just got the paper that day and he looked at it and he said to Rupert Clark, "Look Rupert, we could have written that paper, damn it." He said, you know, but he thought it was great. At the same time, he’s right, of course. They could have written that paper if they’d had the time.
Kellermann
It had to do with luminosity functions.
Van der Laan
That’s right. There were enough identifications to try to extrapolate to actually get the proper volume density relation which they hadn’t done.
Sullivan
Now, what about your CKL? Was that just sort of a straight forward putting together?
Kellermann
Yeah, pretty much. I mean, things just fell right in. I was actually going to go back a bit.
Sullivan
Sure.
Kellermann
I can tell you the exact same story about how I got into radio astronomy. All you have to do is substitute [John G.] Bolton for Ryle and observational radio astronomy for the theoretical. It’s ironic. I mean, you won’t believe it now. We should have done this independently. But I had worked in high energy physics as an undergraduate at MIT. And I went to Caltech intending to do high energy physics at the synchrotron there. I had an assistantship which meant I was supposed to work fifteen hours a week on some research project. And I also was given some weeks to go down and talk to people. And I always had a background interest in astronomy. It never occurred to me that you could do actually this professionally, for a living or anything. But cosmic rays sounded more proper. And I talked to the cosmic ray people and the synchrotron, of course. It was obvious that synchrotron was a loser from the beginning. I mean, there were guys there six, seven years, all doing the same thing even went bouncing [???].
Van der Laan
That was my impression also [???]. It seemed to laborious.
Kellermann
And I talked to a number of people about low temperature physics and everything. And the last one of the list was, they told me that this guy Bolton was doing radio astronomy. He was looking for students and might be worth talking to. And it’s the exact same situation. I went in and there was his secretary sitting there and I asked to speak to Dr. Bolton and they told, "That’s Mr. Bolton over there." And he was right there and he turned around and said, "What do you want?"
Van der Laan
He’s abrupt, isn’t he?
Kellermann
Very abrupt. And I knew about as much about radio astronomy as you did Harry and I sort of said I was sort of interesting in this or something. And he only asked me two questions. One, what did I know about electronics and I told him I was a radio ham and I worked one summer in a microwave laboratory or something. And the other question was how I felt about heights. And what I thought he was referring to was the fact that the observatory was up at an elevation of 4,000 feet, would that bother me. So I said, "Oh, no. It doesn’t bother me at all." What he really meant was how do you feel about climbing on antennas. And that was all he needed to know and the same day, he said, "Well, there is an office down the hall you can use."
Sullivan
Incredible. Well, actually I heard several similar stories, not in these details that are so remarkably similar, but people just wandering in. It seems almost that very few people actually started off always wanting to be astronomers and then looked at this radio astronomy and, you know, conscientiously said this is the way go.
Kellermann
Yeah. I mean I had to work fifteen hours a week and I, even at that time, had no intention of even doing it for a thesis but I had to satisfy my assistantship. And even after this brief interview, all the other people said, "Well, you know, think about it and come back." And it was a gentlemanly way out for both to say, "Well, I’d like to go and look at your record and what not and see if I really want you." But he’d never seen or heard or me but he didn’t care as long as you were physically able and you could dig ditches and plant trees and climb on antenna, you were acceptable. Well, I guess there was a cut off level since you were a Caltech graduate student meant you had some capability or something. And he was just too forceful and dominating to say, "Well, I’d like to think about it." So ok.
Sullivan
Do either of you know if was the standard technique with other students also?
Van der Laan
At Cambridge, most research students were undergraduates at Cambridge so, of course, they would have say a physics lab that was demonstrated by one of Ryle’s people, say John Shakeshaft or John Baldwin, so they’d know a great deal more. There were very few people from outside although Canadians were fairly prominent among the outsiders. You know, people like John Galt and Alan Bridle but also Carman Costain.
Sullivan
Yeah, I’ve talked to him.
Van der Laan
And younger people, yes, like Alan Bridle and Chris Purton, he was there. He overlapped with him. So I don’t know why that was. I think it’s because once you have one contact who goes back to that country, he will send students.
Kellermann
But at Caltech, I mean, the project was just starting. And they needed people so, you know, all students were welcome. He did, as I recall- I was in the Physics Department, you know. I was a graduate student in physics. And Bolton did prefer physicists over astronomers.
Sullivan
Now, did you do anything else in Cambridge besides the spectral work?
Kellermann
No, for a few months were just putting things together.
Van der Laan
It was fun though.
Kellermann
Oh, yeah right. I mean, of course, I saw Harry a lot.
Van der Laan
He used to be across the corridor from me by [?] office. And he’d come in and have an idea and we’d try to shoot it down. Often did but it was a lot of interesting interaction.
Kellermann
I mean, out of that came the first sources with, you know, I mean, the classic source had [?] spectra, you know what I mean. And there were these two sources that had been found by accident at Caltech, CTA21 and CTA102. And I don’t know if you’ve talked to Dan Harris yet. The accident, I mean, in the paper it says they were picked up- It was the Harris and Roberts’ study of 3C sources. They did RA and [?] scans around 3C sources and in the paper, it says, "These few CTA sources were picked up during these scans." Well, if you look carefully, you’ll see that CTA21 or something is about 5° away from 3C86 because Dan fell asleep because he’d been out drinking that night or something. And he let the telescope run for half an hour and ran across this source. And there were these two or three that were found and it was noted...
Sullivan
With inverted spectra, you are saying?
Kellermann
Well, no. It was just noted as previously uncataloged sources.
Van der Laan
[???] comparable to the 3C one.
Kellermann
They subconsciously assumed that it was in the 3C side lobe or something like this or the [?] recorder didn’t work or something. Nobody got very excited about it. And when I went to Cambridge, we actually looked back on the 4C records and there they were. They weren’t strong enough to have been cataloged but they were there and strong enough to pick out and get the flux. And indeed we saw that they had the inverted spectra.
Van der Laan
At first, they looked like thermal spectra.
Kellermann
Well, we didn’t interpret it that way but CTA21 and CTA102 were consistent with a spectrum of mono-energetic electrons. In other words, it was just a gradual roll off at the low end and it wasn’t enough to require self absorption. And then there were these other five or six or so, like Cygnus A and 295, which had clear curvature. And there was also the clear relation, and we put in a separate note in Nature, a relation between the curvature and the surface brightness. The high surface brightness sources had the curvature but we missed the interpretation. I mean, it was a simple-minded interpretation that the high surface brightness sources were younger and therefore they had higher magnetic fields and therefore you had the electron decay, which is not unreasonable but that wasn’t the right answer.
Sullivan
[???] they were compact. [???]
Kellermann
Compact, yeah, but we didn’t realize that they were so compact that they had to have self absorption.
Van der Laan
That was the paper in which you got one of Conway’s students as coauthor. There was some friction, wasn’t there? I mean, Jodrell Bank wanted a big share of the credit. How was that finally resolved anyway?
Kellermann
Well, that’s part of the reason there were two papers.
Van der Laan
[Why there were two papers?], ah yes.
Sullivan
But straighten me out. I’m not clear on this point. How was your [?] CKL superior to, I guess, Whitfield was the standard before that?
Kellermann
That’s right. I mean, it was just another five years later data. The date was more accurate. I think [George R.] Whitfield only went up to 850, his high frequency was the 3C catalog. We went up to 20 and some sources even at 10 cm.
Sullivan
Did it appreciable change the state of affairs though? I mean, you just sort of extrapolated really?
Kellermann
It established more clearly that all the sources had this narrow dispersion in the spectral index, that some of these at least did [?] spectra, particularly the high surface brightness ones, and that there were these few odd cases. We missed the boat in following up. I mean, the fact that these two were discovered by accident should have indicated to do a proper survey.
Sullivan
Well, it’s easy to say in retrospect.
Kellermann
And I finally did just before I left, did this 21 cm survey. It was the so called CTV survey but it wasn’t enough of the sky to get any useful statistics.
Sullivan
OK. Let’s take you back to Caltech then and you were there until when?
Kellermann
I finished in ’63.
Sullivan
Right, but you stayed at Caltech awhile?
Kellermann
No, just for that summer and then I went to Australia on a postdoctoral fellowship.
Sullivan
I see. So you weren’t around for the quasar?
Kellermann
Yeah, that came in ’62. Early ’63.
Van der Laan
February of ’63. We got the [telex?] or something from Caltech [???]. That went through the grapevine very quickly, in a day or two, all around the globe.
Sullivan
And what did that mean as far what was to run up to the telescope and do? I mean, the recognition of the high red shifts, I mean, what did it immediately suggest that you should do observationally?
Kellermann
Well, I mean, the whole program at Caltech was centered around identifications and the collaboration between the radio and optical people.
Sullivan
So, in that sense it proved eminently successful.
Kellermann
Oh, yeah. I mean it was a classic case of setting out a long range program and just stepping along. The first 50 or so accurate positions led to a reasonable number of identifications, ultimately 3C [?]. And the idea was that the small sources would be the one’s position you could measure most accurately. They had high surface brightness and therefore they might be associated with very powerful and distance galaxies. And it was beginning in the ideas at that time that the radio galaxies were not normal galaxies that had emission lines and therefore you could measure their red shifts at larger distance than if they didn’t have the emission lines. So there was this push to identify small [?] radio sources in the hopes that they might be very distance galaxies. And this just went on and on until we got to 295. And then very soon after 295, there was the 3C48 and then 286, I think, which were identified with these stellar objects and these papers were written interpreting their spectra and [?]. And then, I don’t know if you know the story on 273 because that is a little bit lost in the literature. It should have been identified much earlier. I mean, it’s the 10th brightest source or the 6th brightest source in the 3C catalog and it is in a completely clear field. I mean, there is nothing else optically nearby. I think there was just a clerical position in the Caltech right ascension which missed the identification coming to Caltech. Because it was one of the brightest unidentified sources, it was one that John Bolton immediately got onto. He measured a position just with the Parkes telescope despite doing [?] scans and got a position good enough to pick out this very bright 13th magnitude star. The original identification was made on the basis of that, not the occultation. And it was only because of that identification with what appeared to be a bright star that there was all this effort made to do the occultation.
Sullivan
Right. Well, I have talked to Hazard [Cyril Hazard] and I think that’s pretty much what he said also.
Kellermann
And, I guess he told you how they had to saw off part of the telescope to get the occultation.
Sullivan
I don’t remember.
Kellermann
Yeah, the first occultation was at an elevation of some minutes of arc beyond the telescope limits and...
Van der Laan
They dug a hole for it.
Kellermann
No, they had to chop off a little girder or something to get it.
Sullivan
I didn’t quite follow you when you said...
Kellermann
It was because they already had the tentative identification that they knew...
Sullivan
They really wanted to pin down the size, yeah. You said that it was missed at Caltech. You mean when the Cambridge positions came in?
Kellermann
No. Well, it should have come out of the Cambridge positions. I don’t know why. Oh, I know. Cambridge didn’t have it because it is in a side lobe of M87. It has the same right ascension as M87 and with these...
Van der Laan
What’s M87’s 3C number?
Kellermann
274. And with the survey instruments, it was very difficult too because it has a big [???]. But Caltech should have had it with the interferometer and it just got lost in the mud or something. But it was immediately confirmed, of course.
Van der Laan
Did [Sweet?] also get the first spectrum of that or who got first spectrum?
Kellermann
I’m not sure who took the first spectrum.
Van der Laan
[Sweet?] had two spectra, 3C84 and then he had another one to look at [???] but I’ve forgotten which one that was.
Sullivan
I think it was 273.
Van der Laan
[???]
Kellermann
I mean, Schmidt, Greenstein, and Bowen and [?] were all fussing over this stuff.
Sullivan
Ok, let’s move on. You went to Australia. What did you do there?
Kellermann
I went with the intention of carrying on the spectra work, both to do the Southern hemisphere...
Van der Laan
By then, Bolton had left and gone to England.
Kellermann
He left in ’61. So he wasn’t there during most of my thesis work although he got me started. And I went there with the intention of carrying on, both to Southern hemisphere sources and to push it higher in frequency. Remember this question of the interpretation of the extended [?] sources, whether they were really extended sources or components of some others? And it turned out that Peter Scheuer had studied a lot of these at Parkes. I spent a lot of time going through the...
Van der Laan
[Did he do some occultations with the ??] Peter Scheuer did the occultation theory, you remember?
Kellermann
Yeah right.
Sullivan
Right. He did single dish stuff, I think.
Kellermann
That’s right. He did it with the Parkes telescope. And I spent a lot of time going through the original records just to get out spectral data. I think it was done during our first year. Well, there were a lot of sources in the Parkes catalog that had apparently curved spectra and I was trying to measure it more accurately by getting more accurate measurements at a number of frequencies. It turned out a lot of them when you got the more accurate measurements, they were no longer curved. So it was sort of a negative result, not very interesting. And it’s just sort of an interesting thing. There is a side [light?]. I moved up to Parkes from Sydney. And the general way of operation at Parkes was that the telescope during the daytime was reserved for maintenance, every day from 8 to 2. But it wasn’t really used for maintenance. I mean, some days it was but others it would just sit there. And I’d ask to use it to do something and Bolton would say, "No, it’s reserved for maintenance." And the only thing that you were allowed to do in the daytime were those things that could only be in the daytime and that’s the Sun, and Venus, and Mercury. So I put in a request to observe Mercury. And the telescope was just sitting there day after day.
Sullivan
So that’s how you got into...
Kellermann
That’s how I got into planetary stuff. And I mean, the very simple theory we had at the time, you know, that Mercury did rotate meant that you’d only be able to see it when you could see some of the sun lit side. And the big problem, of course, was being near the Sun [in side lobe and everything?]. And the optimum time to do it was clearly at the greatest elongation when it was half lit and you were far enough away from the Sun. Well, now that I had every day to my disposal, I realized that now this is the time to do it when it was away from the sun but it was still very weak. It was very weak and I had to average up a number of scans. We didn’t have computers and everything to do it. And we were able to take the data and they had a scheme where the data could be digitized and the number printed out on an electric typewriter and so every minute of arc or something got printed on a typewriter. And what I tried to do was scan back and forth. And I get the big strips of numbers and then lay them down, glue them down all next to each other and then just add up across to get my added scans. And I realized this was going to be a little bit tricky and it was going to require some practice. So instead of waiting until greatest elongation, I decided to start right then and try to get all the bugs out of it. And the time when I was able the start was near [?] injunction. In fact, the dark side was toward us. And there would be no chance of detected it but at least I’d find out what the solar [???]. I did a couple a scans and by god, there it was, the back side of Mercury, you know, as bright as anything.
Sullivan
Was it brighter than you expected the front side even to be?
Kellermann
Well, I mean, it was [the inferior injunction?] when it was close to the Earth so the flux was large, whereas the expected flux at greatest enlongation would have been a lot less. No, the brightness temperature was about half of what you expected it to be. I mean, you expected the day time side to be around 600 and you expected the night time side to be around 0. And I was observing the night time side was around 300. And then I followed it around for a rotation or so and it stayed constant and it didn’t change. I missed my great discovery then. The moral of the story is graduate students should learn to read foreign languages because, I mean, there were two doctrines about Mercury. One was that did rotate and the other was that it didn’t have an atmosphere and that’s why the nighttime side had to be cold. Well, clearly one of those was wrong. Actually, there had been an earlier observation of Mercury taken at the greatest elongation at Michigan. And they took the standard theory that said if the daytime side varies in the way that you’d expect from solar heating and the nighttime side is 0, they took their observed values as about 400 and said if the backside is 0 then front side has to be 1100. It’s much too hot and they left it at that.
Sullivan
That was the first detection?
Kellermann
That was at Michigan by [William E.] Howard and [Alan H.] Barrett and [Fred T.] Haddock, I guess, and they didn’t have the sensitive to follow it around or anything. Well, I followed it around and I found that the temperature was constant all the way around. It meant that the nighttime side essentially was at the same temperature as the daytime side. There were only two explanations. One, that it really did rotate and the other, that it had an atmosphere that by convection carried temperature around.
Sullivan
Or both as we know now.
Kellermann
Well, no. There is not a significant atmosphere.
Sullivan
Well, I guess that is after this problem.
Kellermann
But George Field had written a paper postulate an argon atmosphere that might be able to do it. And if you look through the literature, the evidence against no atmosphere was a lot weaker against the evidence for rotation because, I mean, there were statements all over the literature about the rotation being equal to the revolution by [a part in a 1,000 or a part in 10,000?].
Sullivan
But that was based on late 19th century drawings?
Kellermann
Well, that was the trouble. It was all the secondary literature that I went to, reviews, and discussions, and textbooks. All the primary literature is in French by Schiaparelli and these guys and, of course, I never went to that because I can’t read French. And it just got escalated. He wrote one paper and people copied that and changed it and what not.
Van der Laan
And I bet less hesitant.
Kellermann
And it just got to be the doctrine. So I picked the wrong one of the two.
Van der Laan
But what did you do, Ken?
Kellermann
Oh, I mean I wrote the paper...
Van der Laan
What did John Bolton say when you discussed it with him?
Kellermann
Oh, yea. Well, we wrote the paper on the observational results.
Van der Laan
I don’t know that paper.
Kellermann
It’s in the AJ.
Van der Laan
Yeah, but what does it say?
Kellermann
Well, we wrote the observation result and what the temperature was.
Van der Laan
You didn’t draw any conclusions?
Kellermann
Yeah, in the end, we said, "It could be due to a circulation in a thin atmosphere suggested by a field or something like that." And then it’s even worse than that, I should have known, because John was away then. He was traveling and he was writing these letters back from everywhere he went. And he wrote back from Arecibo that the Arecibo radar data shows an anomaly in the [?] of Mercury so he said And even that should have been a clue for me to follow it up in that direction but I blew it. So that’s how I got into planetary astronomy.
Sullivan
Was it discovered that early now from the radar?
Kellermann
It was within a few months of that. Back in ’64 [???].
Sullivan
I was thinking it was a bit later.
Kellermann
No, it was in ’64. It was within three months of that. And then of course, for a radio astronomer observing planets is the same as observing radio sources. The Parkes telescope was so much better than anything around and I was able to and I did Venus and Mars. Mars was a hot thing at the time.
Sullivan
No pun intended.
Kellermann
Well, it was just the opposite. The year before Rod [Rodney D.] Davies had come out with this hot temperature for Mars and postulated the Van Allen Belts like on Jupiter. And there was a Mars probe going at that time and they almost called it back off the launching pad and put in some extra shielding and everything to protect against it. So Mars was sort of topical at the time that I did Mars, which is sort of the best understood planet in terms of behavior.
Van der Laan
That gave what you were expecting?
Kellermann
Yeah, 200° or what have you.
Sullivan
So his observations were just mistaken somehow?
Kellermann
It was confusion, yeah. And then Saturn was...
Van der Laan
Were you working at a shorter wavelength than he was?
Kellermann
No.
Van der Laan
He was working at what, 21 cm?
Kellermann
He was working at 21. I did 6, 11, and 21. Venus even at 50 cm to get the long wavelength stuff. And then Saturn was also, I think, Brian Cooper had earlier got the somewhat high temperature for Saturn at the longer wavelengths so I continued that on to even longer wavelengths. And as we know now, in fact, you know that it is a [?] phenomena that as you go to longer wavelengths, you look deeper into the atmosphere and see the higher temperatures. And then we had the first detection of Uranus [???].
Sullivan
This was all down in Australia?
Kellermann
Yeah.
Sullivan
In the period?
Kellermann
’64 essentially
Sullivan
to ’66.
Kellermann
No, ’64.
Sullivan
All in ’64.
Kellermann
It was a little bit less than two actually. I left in the spring of ’65.
Sullivan
Let me ask about those first couple years on the Parkes dish. Did everything go smoothly as soon as the dish was built?
Kellermann
Well, it’s Mark Price who you want to talk to about the first year. I got there when things were pretty well smoothed out.
Van der Laan
Was [Ronald N.] Bracewell’s visit behind you when you went?
Kellermann
The stories weren’t. You know that story?
Sullivan
I’ve heard it. I haven’t checked the actual log book. I’m told you can actually look there to the date.
Kellermann
No, the visitors’ book.
Sullivan
The visitors’ book, right.
Kellermann
Well, an interesting thing connected with the first observation was Bolton tried to make a map of [?]. The telescope was structural complete, could be driven, but the [?] equatorial wasn’t running yet so it could only be run in alt-az coordinates. And he was interested in [?] A which is a great big source. And he sat down, you know, long hand even without an HP35 and worked out the altitude and azimuth directions and what-not to do scans across it and did these scans. But then when he tried to tie them together, they wouldn’t fit and, you know, after a few days or weeks, whatever it was, he gave up. He just couldn’t get it figured out. He hadn’t got the coordinates right. He was just going to have to wait until the [?] equatorial gets in. And the reason he couldn’t get it to fit was that he was using this alt-az dish with fix polarization working his way across the sky rotating and this thing is 30% or 40% polarized.
Sullivan
When was this realized?
Kellermann
Well, it was realized until after the Centaurus stuff. And then after he went back and did it properly then the high polarization was found. It was another one of those things that was there.
Sullivan
You could really see how that could happen.
Kellermann
But Mark was involved in a lot of the early measurements and the discovery of Faraday rotation and everything.
Van der Laan
I see. Was Mark there as a postdoc?
Kellermann
No, he was a graduate student.
Sullivan
Well, I’ve talked to Frank Gardner also, of course, who was involved in it.
Kellermann
Well, there was this business with Bracewell and then they all went away to fight about it and left Mark there. It was over Easter holiday or something. And as he tells the story, the early stuff was done at 11 cm and there was a 20 cm receiver on the telescope. So he went out and did some of these sources at 20 cm and got digital position on them. You can understand that on the back of some physics book or something and try to learn about polarization and ran across Faraday rotation and decided that’s what it was. And then went and did another frequency and sure enough [???]. And so on a weekend while everybody else was away, he discovered Faraday rotation.
Van der Laan
Lovely.
Sullivan
Let’s see, during these times you drop out of the mainstream for a couple years, is that correct?
Van der Laan
No, only one year.
Sullivan
Only one?
Van der Laan
Only one, yes.
Sullivan
[And back to Canada?]?
Van der Laan
That’s right. I finished my Ph.D. in ’63. I guess about the same time you did, July ’63. And then, of course, one could then get any number of jobs and I was offered a number of jobs without applying for them. And I took one that started with a year leave of absence with half salary. And I went to study philosophy for a year which was always my ambition.
Kellermann
Where was that?
Van der Laan
In Amsterdam.
Kellermann
That’s right.
Van der Laan
And I didn’t lose touch with astronomy then. I kept up with the literature and I visited the Amsterdam Observatory and the Leiden Observatory. They had a colloquia there and I got to know Oort and [?] personally. And I’d met [Hendrik Christoffel] Van der Hulst already because Van der Hulst was on leave at [?] Institute at the time of the [?] Symposium that we just discussed. So I’d met him there. They had always been surprise because there is this name H. Van der Laan and I spelled it the normal Dutch way rather than the anglicized way in monthly notices. They never heard me describe the story of the Dutch name and never heard [???] in the monthly notices. So anyway, they thought it was interesting. Anyway, so after a year, I went back to Canada. I found this job that I had accepted beforehand and I spent, let’s see, a summer in Ottawa with the group at [?] and then [???]. But I spent two summers here, ’66 and ’67, and then at the end of ’67- when was [?]? That was in ’66, the URSI [International Union of Radio Science] thing in Munich. Did you go to that?
Kellermann
No, I didn’t go to that. ’66, that’s right.
Van der Laan
I went to that. It was just after I’d written that little note in Nature on the extragalactic variable sources. And ’67, I was here and as I came the [Apr/Art?] paper came out. So I thought [?], he’s right, we have to do something completely different so there is no point in me doing any problems unless I’ve convinced myself that I was wrong. So I sat down and convinced myself that I was wrong. And I was completely open minded about it. When I came here and read the paper, it seemed so plausible. I thought probably the guy is right so I’d be wasting my time unless I settled for myself whether he’s right or not. And I convinced myself he was wrong and I walked into Dave [David S.] Heeschen’s office one day and said, "[Arp/Art?] completely wrong." "Jeepers, let’s hear about it," Dave said. He was rather interested. Dave had just been out to a trip to Harvard or something. Everybody was talking about this [Arp?] paper in the May 5 issue, I think, of the ApJ. So I convinced Dave and Dave said, "You better write that up." So then we improved things a bit and Frank [?] wrote a computer program. [?] and I wrote that paper.
Sullivan
I see. This paper, I’m not familiar with this. What is this?
Van der Laan
That’s the paper that shows how the [Arp?] effect off pairing can arise in the selection effect because you show only what you find and not what you’d look for but you’ve not found. It’s really a confusion of a priori and a posteriori probabilities. And we verified him empirically by generating phony [?] catalogs. And, in fact, John [Bacall?] had taken this paper [over?] in that recently published debate between him and [Arp?].
Sullivan
I see. It’s in that collection.
Van der Laan
He reprinted in there, yes.
Sullivan
And who are the authors again?
Van der Laan
Frank [?], [???], who was here. So that’s what I did here and then [Blaaw?] offered me a job at Groningen and Oort offered me a job at Leiden. And although the job in Groningen was [?] and the one in Leiden was not, I went to Leiden because I sort of thought the prospect of working with Oort was rather nice.
Sullivan
This was in ’67?
Van der Laan
Yes, and my wife wanted to go back to Holland. That’s why I was looking for jobs in Holland in the first place. I didn’t. I was only faintly aware they were building some radio telescopes.
Sullivan
I see. That wasn’t a part of your motivation at all?
Van der Laan
Absolutely not. I didn’t want to go back and I always thought we would go back to Canada or the U.S. I thought I could never get used to again the Dutch way of life. But then I enjoyed it and I [?] into Westerbork in Holland and finally even got a chair when Oort retired, which I’d never dreamt about. That was certainly not the reason for going back to Holland. I wasn’t even aware how old Oort was.
Sullivan
Well, there is a lot of interesting stuff after this point but, I guess, I’m cutting it off at about ’65 or so because it just gets out of hand. But I’m making a couple of interesting incursions like pulsars but I’d like to ask you about your ideas of the beginning of the VLBI. You were in on the American end, weren’t you?
Kellermann
Yeah. That came out over a pitcher of beer. It’s probably not surprising, continued over pitchers of beer. The Canadian stuff, I think, went back a fair number of years.
Sullivan
Well, I have talked to John Galt before. He’s told me his version.
Kellermann
Nothing was done in the U.S. It was the summer meeting of the AAS in ’65 in Ann Arbor.
Van der Laan
Yeah, I was there.
Kellermann
Pardon me. Were you?
Van der Laan
Yes.
Kellermann
Yes and it was just Marshall Cohen and I and I don’t remember who else at a big table. Lots of pitchers of beer and we got to talking about it.
Van der Laan
Was [Alan ?] there?
Kellermann
No, no. As I said, it was quite independent of the Canadian project. We first found out about that during that winter. But, I think, a lot of people had thought about this independently earlier, many years earlier, as early as 1950 or ’52. I mean that’s what the Hanbury Brown interferometer was invented for. And they thought that radio sources were stellar and you needed transatlantic dimensions to resolve it and that’s why he developed the intensity. I mean, he had no way of keeping coherence then to develop intensity in the interferometer and the concept of taking tape recorders.
End of Tape 35B
Sullivan Tape 36A
Sullivan
Well, I’m not familiar with this Jupiter...
Kellermann
In ’64, I guess, they started doing intensity interferometry on Jupiter using tape recorders. I think...
Sullivan
Smith...
Kellermann
That’s right. Smith and [Carr?], they were getting the [?] from WWV and you can just use a tape recorder like this and a few kilohertz bandwidth. Jupiter was so strong it was no problem. But then, a lot of people, I think, had probably thought of it earlier. And then just around the mid- 1960s, all of a sudden you had tape recorders that could give you another bandwidth to detect extragalactic sources and commercial frequency standards, which were good enough to maintain coherence for at least some tens of seconds. And we realized that you could do it, I mean, with existing equipment. And the really big advantage which we had, which no longer exists at NRAO or probably anywhere, I remember going to Dave Heeschen in September or something and spending about 5 minutes total describing this thing, you know, what it would take to do it. And he sort of opened up a book and said, "How much will it cost?," and I said, "It looks like it would be about $100,000 to develop two recorders and everything." He looked in a book and this is in September and he said, "We can’t spend more than $50,000 on it during this year," you know, up until December, "Will that be enough until December?" and that was it. And a week later, I saw him in the hall and he said, "What have you done? You haven’t spent any money or anything." And if we had to go through ordinary channels of writing proposals and applying for grants and what not...
Van der Laan
Or university administration.
Kellermann
Yeah. It would have been years. Right. I mean, NRAO had this budget for new projects and new ideas and what-not and there it was.
Sullivan
This is late ’65?
Kellermann
That’s right. And then we got aware of the Canadian project and talked to them and everything. We went pretty different directions because they had these surplus TV recorders and they did analog recording which is much more difficult but they ended up with a larger bandwidth. We initially had the digital recording just using computer tape drives which was much more straight forward although in terms of [???]. In other words, they had to have their tapes aligned to within a microsecond or so but we didn’t because we could just buffer the data [???].
Sullivan
And look for fringes?
Kellermann
Yeah.
Sullivan
And so the timing was such that they got fringes first?
Kellermann
By a few weeks, yeah.
Sullivan
Only by a few weeks?
Kellermann
Oh, yeah. Let me see, well, we were too ambitious. Our first experiment was between Green Bank and Arecibo and there were no results. And their experiment was between [Penticton?] and [ARO?] or something and I guess there were no results on that. And they dropped back to a baseline between Ottawa and some suburb of Ottawa or something. It was, I don’t remember, some 10s of kilometers 100 kilometers or something. And we dropped back from Green Bank to Maryland Point. And they got fringes on their short baseline a few weeks before we got fringes on our short baseline. And then they immediately went out to [Penticton?]. And we were both at that time much too conservative about how high in frequency you could go with the existing frequency standards.
Van der Laan
That was in ’66 or so.
Kellermann
That’s right. They went to 75 cm. We stuck our necks out to try 50 cm. And then we quickly jumped to 18 and did stuff from Green Bank to Haystack and it was all continuum. All the stuff had been built up for continuum work. And then about the same time all the [?] OH lines were discovered and the maser concept and the idea those should be very small.
Sullivan
Who was the one that really pushed the OH VLBI?
Kellermann
Jim Moran.
Sullivan
Was it him, not Burke?
Kellermann
Well, the two of them, I guess. I mean, he was a student of Burke’s but he did all of the work. He wrote all the programs and everything to analyze spectral line data.
Sullivan
Did you realize in those first experiments that the technique would prove so powerful, I mean, worldwide and all this sort of thing?
Kellermann
Yes and no. We wrote and Gold wrote a paper on these various other applications. I had a mental block and it was partly because of this, you know, expanding source model, which we believed by the gospel, that all we had to do was measure the size of the source. All you need is a few points and the visibility function. They would be spherical symmetric. And if we could do as a function of time, we would actually be able to measure the expansion and everything. So we did Green Bank to Haystack. And then we went out to Hat Creek and that was 20 million wavelengths cross country at 18 cm. And only at a few points [???] you can interpret the stuff as circularly symmetric. And then we were going to go to 6 cm at Hat Creek. That was in the winter of ’67.
Van der Laan
’67-’68.
Kellermann
’67-’68.
Sullivan
But why were you believing that when there was quite a bit of evidence for double?
Kellermann
No there wasn’t, not at that scale. No, that’s what I’m trying to say we should have. We should have looked back. I mean, we were aware of the conventional interferometry, all those sources were doubled.
Sullivan
On that scale you didn’t.
Kellermann
Yeah, but they went through the same stages. In the early days, you had one point (phrase?) that gave a size to the source and, of course, Cygnus was discovered to be doubled and everything. We didn’t take that hint. As I said, I had this mental block. I just told myself they had to be single.
Van der Laan
[???]. I never did. I published in a way to stimulate observers so they would have curve on which to plot their points. It was just a two coordinate plane in which to plot their points but stimulate them to go and observe and to knowing meaningfully how the observations should be spaced in frequency and time. I never expect any source to conform to it.
Kellermann
Yeah, but you believed then and I did, and I think you and I do now, the expanding source model, not quantitatively expanding in the uniform velocity (phrase?).
Sullivan
The basic concept, yeah.
Kellermann
The basic concept.
Van der Laan
Yes, that’s right. I mean, [of the dimensions increasing, increasing time?].
Kellermann
Right, the opacity moving down in frequency that was in our papers and you paper. We were going to go from 18 cm to 6 cm cross country in that winter and at the last minute, Hat Creek backed out because they said they didn’t want to change receivers in the middle of the winter because they might not be able to change back to 18 cm till the spring or something. And by coincidence, the Swedes had decided they wanted to get involved it that stuff and Burt Hansen...
Sullivan
By then because of the maser sources, I suppose?
Kellermann
No, yeah. No, they were just interested technically, I think, technically and [Olef E. H.] Rydbeck on the spur of the moment sent Burt Hansen to Green Bank. And Hein Hvatum, of course, had been there and knew him well and everything. So he came over here and sort of said they were interested and would like to try an experiment sometime. Well, we just within a couple of days had told them that we couldn’t do this thing, we had plans with Hat Creek.
Van der Laan
[???] an 85 foot dish?
Kellermann
Yeah, yeah. We couldn’t do this thing with Hat Creek and said, "How about next month?" As it turns out, I was getting married in Holland then and I was going to be in Europe.
Van der Laan
[???]
Kellermann
Yeah and well, we did have a real one first but then we went off to Sweden [???]. Our previous thing had been 20 million wavelengths from here to Hat Creek at 18 cm. And we went from here to Sweden at 6 cm which is 100 million wavelengths. But again we found the smaller and smaller sources but we were still just putting a size on them and everything. And it wasn’t until ’70 or so that the data finally began to strike home the point that the sources were more complicated than that. And then we first began to realize that you needed lots of spacing and lots of antenna pairs and full hour angle tracks and what-not.
Sullivan
Just like the arc minute resolution. I just saw [???] which I’m sure both of you could make a few interesting comments and that is source variations. [Dent?] in ’65 was the first to pick up on any but how did this affect how you thought about radio sources? Was it particularly surprising?
Kellermann
Well, because I had been playing around with self absorption and everything, if you read [Dent’s?] original paper, he interpreted it incorrectly. Let’s see if I remember, he said that if it’s varying and it was at such and such a distance, the size had to be less than the distance light could travel during that time. And therefore it had a certain angular size and with that angular size it should be opaque at that frequency. And this was at 4 cm, it had the same flux at 4 cm as it did at longer wavelengths and therefore it wasn’t opaque. And what he didn’t realize was that the variable component was opaque and he also had a numerical mistake in the calculations. George Field pointed it out in the paper. So in a way, I believe it more than [Dent?] did, that they really were compact sources. See, I should have said that before. That’s where the whole idea of the VLB stuff came in too. Immediately you had the indication that there were sources of 1000th second of arc and that’s how I got interested in the VLB stuff.
Sullivan
But was it particularly surprising? Was everyone sort of expecting that sooner or later these sources, they would be found to be varying?
Kellermann
No, I don’t think so.
Van der Laan
I don’t think so. I was completely astonished that extragalactic sources at that distance, hundreds of millions of light years, could change in the course of a man’s lifetime, let alone in a month or a week. It was a wholly new concept. If you think right back to Greek culture, that is something that shouldn’t change, you know. And certainly if something could change on a stellar scale and we knew that you could have a supernova event in a galaxy but that such things could change. Now, we take it completely for granted as we do so many stupendous things. But ten years ago, no.
Sullivan
Well, it would be the same shock now if we saw a whole galaxy change somehow, the optical of a whole galaxy.
Kellermann
That’s right. But Ivan [I. K. Pauliny-Toth] and I set out- the 140 foot was just finished then and we set out to follow some of these. And then there was this tremendous outburst in ’66, in the spring of ’66, and in several other sources. And we had observations at several wavelengths and I mean, saw this effect of the peak flux occurring at longer wavelengths at later time.
Sullivan
That was the first outburst also that first showed that effect nicely?
Kellermann
Well, it’s phenomenal. There were three outbursts at that time in 273 and 279 and 454.3. Nothing that has happened since then is comparable to that. And 3C120. There were three outbursts or two in ’66 and ‘7 or something and they all showed this effect. The sources were opaque. It was clear that you could interpret it, at least qualitatively, in terms of an expanding source. And then Harry showed up and interpreted it quantitatively.
Van der Laan
At first, we were actually planning on writing a joint paper putting in the theory and the observations. But then observations always take longer [???]. Fred Haddock and Geoff [Geoffrey R.] Burbidge were on the visiting committee of NRAO and they visited here just after I got here. Fred Haddock talked about how he didn’t understand a bit of it at all and then I said, "Well, Ken and I are planning something on this because we think it may have a very simple interpretation. Should we write it up?" But finally it was clear I could only be here six weeks because I had commitments after that to do something in Canada and Ivan and Ken were so busy. I don’t know, you were going off to some conference [?].
Sullivan
This is the summer of ’66 after [???]?
Kellermann
The summer of ’66. No, just a vacation, I think.
Van der Laan
Whatever it was. Anyway it turned out to be impossible to write a joint paper incorporating both the data and the simple model. So we decided that I would write up the simple model which was terribly simple. It’s just the generalization of [Iosef Samuelovich] Shklovskii’s 1960 supernova remnant one, nothing else. It’s just [???] and I honestly did it with the motivation was mainly not that I think this is the way the world is but if we do this, because Ken and I both realized that we somehow need many simultaneous observations at different frequencies and people have to be really convinced and understand what they are doing or they won’t join this bandwagon. And this thing needed a bandwagon because observations had to be made simultaneously and were inherently non-repetitive. You couldn’t confirm your observation afterwards. And realizing all that, we said, "Well, we better write it up." So I start sketching this thing. Robin Conway was here and he said, "Yes, you better write it up." And I wrote it a draft and Bill Howard said, "Yes, you should really do this." Bill said, "I know from experience how this helps, this stimulates an observer to have to done something." It’s like Ryle having something to shoot at. But an observer would be just as pleased to disprove a model as to confirm it.
Kellermann
More pleased.
Van der Laan
Well, you needed a model. That’s the point.
Sullivan
Well, you were playing that role today with the radio source models.
Van der Laan
Yes, that’s right.
Sullivan
What was the year that you said the extrapolation came out to the end of the world?
Kellermann
No, it was 2025 or something. 3C273 was going up 10% a week or something.
Sullivan
Ok, well thank you very much gentlemen.
Van der Laan
I should tell you one more thing.
Sullivan
One more thing, go ahead.
Van der Laan
Not about my own work but I sometime think there is a very big discrepancy between how outsiders feel about the Cambridge group and how we experienced it as insiders. I mean, I only became aware of the Cambridge [?] reputation in the outside world after I left the group. Before I went into the group, after all, I knew nothing about radio astronomy so I certainly had never heard of the Cambridge group, let alone any reputation about it. I joined it. I found that Ryle was clearly the dominate personality. I don’t only mean that he was the strongest character. He was cleared the most widely interested and enthusiastic man and inside the group, there was a genuinely fruitful research atmosphere. I mean, the group was in general extremely open. We talked about everything and Ryle was and that’s quite a contrast with many people I know in that kind of position.
Sullivan
Well, they all admired him also [???].
Van der Laan
Many people actually loved him. I was and still am very fond of him. Some people didn’t like him and he didn’t like some people and he was very strong in his likes and dislikes. If he didn’t like you as a research student, you were in a very tough spot. He would really discriminate against you. But we got along famously and well, I’m the kind of chap where if I have an idea I have to discuss it with someone. And I often walked into Ryle’s office and interrupted him and he would listen and be genuinely interested. And if he just couldn’t spare the time, he was obviously pained that he had to write some report or a letter because he would much rather listen to you. He conveyed that impression and I think it was authentic and that a tremendous stimulus. I mean, he was genuinely interested and enthusiastic. And I only realized after I left the group that the group had a somewhat narrow minded attitude towards the outside world. And in retrospective I can see that maybe I was just naive and could have detected that earlier because I know that Ryle did sometimes explicitly warn us that if some visitor were coming in- I remember once Mort Roberts was visiting us- that we shouldn’t give any information away. Although he did say explicitly and I think that’s rather nice, that, "Look, we had very little money." He once said to me, "NRAO has more money for its cafeteria this year than we have for telescopes." And he said, "Now if this visitor is not capable of asking intelligent questions. Don’t give him the answer to questions he should have asked. Ok, but give him honest answers to his questions but don’t volunteer a great deal of information because we have to be clever because we’re not rich." And we understood that. We accepted that. Only in retrospect did we understand that this really often took on pretty extreme forms.
Kellermann
Yes. I’m not sure it was so bad. I didn’t realize it was that bad. When I went there, I think, there wasn’t this cloak of secrecy. At least, I was fairly well accepted. I had a key to the place and worked nights and everything. Well, the most fascinating thing was these Saturday morning sessions in which they would rip each other apart. Well, first he would go through, the people there had their assignments for the week or something they were supposed to do a report on what they got done and each one got ripped apart as why you only got 1/10th as much work as you were supposed to do. And then he would get onto the literature discussions.
Sullivan
Who made these assignments?
Kellermann
Ryle.
Van der Laan
No, John Shakeshaft kept that as a job. So there was a weekly output of new papers to be reviewed and behind every paper was a name of a research student or a junior staff member.
Kellermann
No, I’m talking about the assignment to the observatory, like fixing this antenna.
Van der Laan
Right, right.
Kellermann
And then came the literature part.
Van der Laan
They were also assigned. All the papers were assigned. John Shakeshaft did that and some students wouldn’t do anything and others would say, "Ok, I’ll review my things this week." The question would be, "Who’s got a paper to review?" There’d always be something to review. Often this would take all Saturday morning but it was very interesting. And Graham Smith and Martin Ryle would cut into you if you didn’t give a good, critical review and if you’d missed the essential point [???].
Kellermann
He gets very emotional too. I saw cases where some poor student would get up and merely report on what somebody else had said in his paper and something that was obviously wrong. And Ryle would jump into him and criticize it as if it were his theory or his...
Van der Laan
As if it was his mistake.
Sullivan
Yeah, that’s easy to do. John Bolton has told me about these Saturday morning sessions and how awful he thinks they were.
Kellermann
And poor, old [George C.] McVittie came once and reported on source counts from VRL based on 30 sources and then went on to say something about the MSH catalog and he told him, "Oh, that stuffs all wrong. Forget it. Don’t even pay any attention to that catalog." It was a very lively place. Aside from doing that spectral work, I got an awful lot of experience in that three month period. Really a lot of enthusiasm.
Van der Laan
The place was really alive. And at lunch and at coffee and tea time which we would have in the afternoon at Cavendish, always talking astronomy. If we didn’t have a piece of paper, we would spill some of our tea or coffee and we’d draw with tea or coffee on the table. But it was just very enjoyable. So I mean, in retrospective and all at the time, I really enjoyed those years and I think as performative years of a research science, they were terrible important. And Leiden Observatory, we tend have an atmosphere that is rather akin.
Kellermann
Well, I don’t know. When I spent time there, it was exactly the opposite. It was the dullest place I’ve ever been. Just before you came. We traded offices, remember?
Van der Laan
[???].
Kellermann
I stepped out of the office and you stepped in.
Van der Laan
That’s right. When I came in, that’s right. No, I think in the last few years, Leiden has had an atmosphere that...
Kellermann
I gather it has but it was different. It was like a morgue.
Sullivan
It’s not a very define research group. I mean, you have many different fields.
Van der Laan
Well, it’s broader. It’s an observatory rather than a radio observatory.
Sullivan
Well, even Cambridge, it’s not a radio observatory like here. It’s radio sources and a few other things.
Van der Laan
I know. That’s why, in part, maybe because they are so specialized. They also feel like own a monopoly on it. They own Cygnus A as it were. [We can’t have anybody else doing anything on it?].
Sullivan
I’ll cut this now
Sullivan
Afterwards, Ken Kellermann brought up the very interesting part about why it was that Caltech missed the opportunity to do aperture synthesis. Because they had the two steerable dishes and an interferometer and they were the only ones that had such a set up as early as ’59 or ’60 and yet they did not make the connection of doing Earth rotation synthesis. They did various baselines. They had 10 stations there. They took various position angles in the UV plane but they never just quite put it together. And all this time, Ryle was fighting like mad with the Science Research Council to get funding for the One Mile Interferometer which would be able to do this. And he was able to keep it secret because, of course, if the idea had ever got out then the Caltech people would have done it right away. Kellermann speculates that perhaps it was the crystallographer at Cambridge that made the radio group just think in the Fourier plane so that it was just sort of a natural thing for them. I think that it was just that Ryle was clever and even from the mid-50s realized the power of the method. Ryle at one time apparently said that the whole Cambridge budget was only equal to the NRAO cafeteria and that therefore they had to be much more clever. And when there were visitors and they asked you a question, don’t give them more information that the quality of the question implied but to answer any questions straight. In other words, don’t give out any more than you need to and from that somewhat the tightlipped reputation has grown. Well, this end the interview with Ken Kellermann and Harry Van der Laan on 19 March ’75.