Interview with Antony J. Hewish
Description
Creator
Rights
Type
Interviewer
Interviewee
Location
Original Format of Digital Item
Duration
Interview Date
Interview Topics
Notes
The interview listed below was originally transcribed as part of Sullivan's research for his book, Cosmic Noise: A History of Early Radio Astronomy (Cambridge University Press, 2009). The original transcription was retyped to digitize in 2016, then reviewed, edited/corrected, and posted to the Web in 2016 by Ellen N. Bouton. Places where we are uncertain about what was said are indicated with parentheses and question mark (?).
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
Begin Tape 60B
Sullivan
This is talking with Antony Hewish at Cambridge on 19th August 1976. I think you came here as a research student originally, is that right?
Hewish
Yes.
Sullivan
And when was that?
Hewish
That must have been 1948.
Sullivan
Coming in as one of the very first ones, how was it determined as to what you worked on? Was it simply Martin Ryle said, “Well, we’re doing this and - ?”
Hewish
Yes. To begin with, I had charts and I was cutting them up into days and nights, actually (laughter). But I don’t think I did that for more than a day. About the very first job I had was designing antennas for the first survey – radio star survey. When I got there, radio stars, as they were called, had just been discovered – Cygnus and Cassiopeia, were known about. The next step was to have an antenna which was a bit more directional and just make the first real sky survey. The first job I ever had was to design an antenna array to do that, which was a broadband array at 80 MHz.
Sullivan
This is what became the Long Michelson - ?
Hewish
That became the Long Michelson, yes, in fact. And I designed the, oh, I got the dipoles and the feed system going for that. And that, of course, was just stooge work in the sense, you know. I was a new pair of hands – this thing badly wanted doing and I did it.
Sullivan
Did you have any radio background?
Hewish
None at…Oh well, I was in radar during the War.
Sullivan
Oh, that’s right. You came from RRE.
Hewish
RRE. RRE and TRE actually. So I knew about electronics and so forth.
Sullivan
All of that early group came from TRE didn’t they?
Hewish
That’s right.
Sullivan
Were there any new principles in this Long Michelson?
Hewish
None at all. It was just absolutely standard, really.
Sullivan
And did you have any idea of how many radio sources you were going to turn up, or radio stars?
Hewish
Not a bit. No, that was absolutely unknown.
Sullivan
So what happened when you turned it on, so to speak?
Hewish
Well, I didn’t actually turn it on. I mean this was very much a group effort. It was going at the same time as I was doing my own programs. You don’t always design equipment for yourself in this group. And that was the thing I did at the moment I came. But, in fact, the first research I did was on the sun.
Sullivan
In fact, I was thinking you were a co-author of the 1C, I’m sorry, you’re not.
Hewish
No, I’m not a co-author of 1C.
Sullivan
So you helped with that?
Hewish
Oh yes. I designed – I did a lot of design on the antenna but I didn’t, in fact, do the work with it. I was concerned, in those days, with, well very initially, with the sun. The radio sun was the most rigorous active radio source known. And it was a period of fairly intense sunspot activity. And I did some work on the polarization of the rapid bursts. I designed a polarimeter and did little work on that. It never got published because it wasn’t terribly interesting and it got overtaken. My interests then went to the fluctuations. And after about a year I was deep into the fluctuations story.
Sullivan
Which was ’49 or - ?
Hewish
Around ’49, yes. I was analyzing, in fact, the thing that triggered me off on that was that I was analyzing records taken with the – let’s see now, it was records taken with 80 MHz array which was doing the sky survey. And it was just running records day after day. And I had to – I was analyzing these for the fluctuating signals on them. We weren’t sure then what they were. They could have been ionospheric or they could have been intrinsic to the source. We didn’t know. The sources were coming through with sidereal time, of course, marked on the recordings. And I noted that afterward – looking back over a run of several months that these things actually – the scintillations were only happening during the night. And I was the first person to actually notice that. And that really got me interested in the scintillation story. From then on, really, I was concerned more with the ionospheric fluctuations. Well, I mean they turned out to be.
Sullivan
You were looking at these records to investigate scintillations?
Hewish
Oh yes, right. I was looking for daily variations, actually, of it, day to day. But what I noticed was that when you had a year’s worth of data to look at as the sources slipped around in solar time so they scintillated during the night. And, in fact, the diagram that comes in that first Ryle and Hewish paper – that was where it was all published – those diagrams were the ones I plotted soon after I came here. And that really got me interested. I had done something for myself, here was a new result and I got quite interested in it.
Sullivan
I’m a little confused as to what the general feeling was. Hey’s original discovery of the Cygnus fluctuation was ’46. And there is a period in there where, looking at the published literature, it’s difficult to understand what people really thought about these fluctuations. Can you - ?
Hewish
Of the fluctuations? Yes, well when we started we didn’t know what it was. And it was really the long baseline. It was comparing fluctuations here and at Jodrell Bank and actually also at Defford. I don’t know if that’s got into the literature anywhere. But there was a paper with Graham Smith’s name on it and I think Lovell’s name on it, there may have been somebody else.
Sullivan
There were two papers together - ?
Hewish
Two papers together, yes.
Sullivan
Smith and then Little and Lovell.
Hewish
Little and Lovell, that’s right. Well I was in sort of – it was the diurnal variation. I discovered just about that same time as that. And, of course, that really clinched the ionospheric origin. And from then on, I tried to – I mean my interests then was to see what we could learn about the ionosphere. Given this new technique. You see, it wasn’t an ionospheric group, I mean, you understand that in the Cavendish it wasn’t just radio astronomy then. It was Ratcliffe. Ratcliffe with the ionosphere, and so on. I had a background of ionospheric physics and my immediate interest was to catch radiation that’s gone right through the ionosphere because this give you data which – just pure reflection techniques don’t.
Sullivan
Just a very good probe of the whole thing?
Hewish
Yes.
Sullivan
You say you had background in ionospheric physics?
Hewish
Well, I mean Ratcliffe’s -
Sullivan
Taking courses around the University - ?
Hewish
Yeah, that’s right, just the ordinary graduate courses, and third year physics courses.
Sullivan
But I don’t think you’ve still answered – was there a raging debate for a few years as to the nature of these things or was everyone just throw up their hands and say, “It could be intrinsic, might be in the ionosphere?”
Hewish
Oh it was 50/50. It wasn’t a raging debate. I mean it wasn’t really that exciting. I don’t think people were really hopping up and down about these things. But I remember when Ratcliffe said this must be the ionosphere. We rather joked at him and said, “Well, you know, he can only think about the ionosphere. It’s probably astrophysical, but in fact, you know we laughed a bit about that. But he, in fact, was right.
Sullivan
So this is, in fact, what got you off into this whole business - ?
Hewish
Of fluctuations. I mean, I got onto the theory after that. Absolutely. That straight on – I’ve forgotten -
Sullivan
You just moved on to one medium – I’m waiting for you anytime to get into intergalactic medium of fluctuations -
Hewish
That’s the last one (laughter).
Sullivan
Well maybe fluctuations through your background -
Hewish
Yes, something like that.
Sullivan
As you say, Ryle and Hewish in ’50, and then you published a paper in the Proceedings of the Royal Society -
Hewish
That was a theoretical one.
Sullivan
’51?
Hewish
Yes.
Sullivan
And using what’s a common model now – diffraction screen model – was this entirely new at that time?
Hewish
Yes, it was new. I noticed the link-up between the diffraction theory and modulation. I mean, I think I was the first one to really know how a phase screen operated and get some quantitative information out of it.
Sullivan
There was similar work going on in Australia, at least from an observational point of view. Did you know about that work?
Hewish
Oh yeah, we knew roughly what was going on down there. But their thing wasn’t so much a nocturnal scintillation. They always had a slightly different story to us. They had daytime scintillation and my first theories on this were that it was to do with the ionospheric fluctuations called “Spread-F.” There’s a thing called “Spread-F” which is something you get on the “p-f” reflection records. Do you know what I’m talking about?
Sullivan
Well, I’m sure of the term -
Hewish
Yeah, it’s simply a multiple echo essentially from the F-region.
Sullivan
(?) frequencies?
Hewish
No, no. This is just the ordinary sweep frequency technique where you send pulses up and you gradually increase the frequency. And you get what is called a “p-f” record. When it penetrates you just get long delays and the record kind of goes along and flips up.
Sullivan
That’s what “spread” refers to?
Hewish
And the spreading is actually a sort of multiple reflection height.
Sullivan
I see.
Hewish
The end of it – if I can plot “p-f” for you – looks something like that… and then you get the ordinary ray and that’s the extraordinary – something like that. And just causing delay - can’t remember what these darn things are now - what is “p-f”? This is frequency and p- is, well, it’s really delay, I think. And the delay increases as you penetrate because you get the infinite – you get the imaginary refractive index business. Well this thing just spreads up and you get a fuzziness like that which is called “Spread-F.” And I found a very high correlation between scintillation and Spread-F, which told us it was an F-region thing high up, you know 300km. And you said, “What were the Australians doing?” – they were having daytime scintillations as well and they thought it might have to do with the E-region. They could well have been right, because I think you do get that down in the Southern Hemisphere.
Sullivan
I see.
Hewish
But at any rate, that was the sort of – there were involved with it. But the difference was they were looking horizontally really. They had this different interferometer – they were looking over the sea and we were looking at the zenith. And we thought they could have been upset with low-angle fluctuations and things of that sort.
Sullivan
It really made it much trickier, I would imagine.
Hewish
(?) it made it more difficult to sort the thing out, yeah.
Sullivan
Looking at the next publication you have in Proceedings of the Royal Society in ’52 in which you had two aerials at 1km apart.
Hewish
That’s right.
Sullivan
So this was really an extension of the Smith and Little and Lovell - ?
Hewish
No, not quite. Well yes, in a sense, but it was done differently because they were just looking for correlations on a very wide baseline to see if there was any agreement. Because if it was intrinsic. The correlator was (?) a kilometer, of course. My interest was to try and measure ionospheric winds and to get the actual structure of the diffraction pattern. I mean once we knew it as the ionosphere it was probably a very small scale pattern on the ground. And my idea was to see correlations and actually get the drifting pattern and learn something about the ionospheric winds, which I did. I think I was the first person ever to measure an ionospheric wind, actually using scintillation techniques.
Sullivan
I don’t have it in the abstract – velocity - oh yes 100 and 300 meters per second.
Hewish
Yes, that’s right.
Sullivan
Height of 400kms - Was this term ionospheric wind only being introduced then or - ?
Hewish
Well, people were studying the ionosphere by reflection also. On getting this thing. People like – does the name Briggs mean anything to you?
Sullivan
No, I’m afraid I don’t know.
Hewish
Well Ratcliffe and Briggs and others were getting correlations in the reflected wave and getting ionospheric winds out of that. And I just extended the same idea really, essentially to the transmission through it.
Sullivan
So it’s really not in terms of your published work anyway until you get to the solar corona occultations that you began doing astrophysics as opposed to ionospheric - ?
Hewish
That’s right.
Sullivan
Well I guess that is the next publication.
Hewish
Although while this was going on, I mean, we’re very much an integrated group here, as you may have realized. And I was doing design work for radio telescopes all the time. I mean designing dipole arrays and things while this other work was going on. But you don’t write papers about how you design antennas, you know. It’s just the astrophysics that comes out.
Sullivan
And you weren’t concerned at all with – like you say I have definitely seen that and everyone I’ve talked to is that everything is chipping into everyone’s problems. So were you involved, for instance, in what the nature of these radio stars was?
Hewish
Oh absolutely, yeah. I mean, regularly we just have a morning session, Saturday morning usually, just sitting around on the floor saying, “What the hell does it all mean?”
Sullivan
Was this before the formal one began – more formal - ?
Hewish
Well they generated from that. But, I mean, the literature was very easy – literature coverage was no problem in those days, you know. One Nature letter every two weeks or something. And the literature was about that thick so that you could really get down to it. And each new result that came out we all got around and talked about it.
Sullivan
Was there anyone else in Ryle’s group working on ionospheric problems as much as you were?
Hewish
No. I was always a bit of a loner on that, really. I sort of had these two interests, the astrophysical ones and the ionospheric.
Sullivan
How did the idea of the Taurus A occultations come up?
Hewish
That came up - There had been a predication that if the sun behaved like a spherically symmetrical ionized cloud with the density decreasing outwards that you’d get geometrical optics situations so that radiation comes down and gets deviated like that. And if you’re observing somewhere down here the radiation cuts off. The sun looks like a diverging lens. And what you would expect to get under those conditions as you track a radio source pass behind the sun would be a sudden cut-off, a long way away from the sun where this caustic curve intersects your line of sight. And that radius – what was called the occultation radius – would have given you the electron density. Now but the records didn’t come out looking like that a bit. The first two times it was tried, I think, were failures because the sunspot activity and so forth – this was the sort of experiment you do every year for the few days that the Crab Nebula is going past the sun, you see.
Sullivan
Which year was the first year?
Hewish
Oh, it was around ’51, I think, or something like that. There was a Machin and Smith observational paper in Nature which actually showed the curve – it wasn’t – if you plot the intensity as a function of time what this would give you would be an intensity which cuts off very sharply like that. What was actually observed was something much more like that, much shallower. And the suggestion was thrown out that if the solar plasma wasn’t varying smoothly but was, in fact, a bit ragged then this thing would be smeared, you see. Well, of course, I got interested then because that was the same thing as the ionospheric scintillations as far as I was concerned. Except the parameters were widely different. But, I mean, it’s basically the same phenomenon. And I then interpreted this in terms of the scattering model. And then went ahead and checked that by seeing if these curves – if the actual scattering angle went to the square of the wavelength. And I did some interferometer records for some years.
Sullivan
So you were not actually involved in the first Taurus A occultation?
Hewish
No, no.
Sullivan
Like you say, you got interested in it.
Hewish
Yeah. No, I mean the result was actually Machin and Smith who did the first observation of that and showed that it didn’t come out looking like it should have done at all.
Sullivan
But now I see that in the – your article in ’55 in the Proceedings of the Royal Society that you talked about irregularities in the sun which was – was that their original suggestion on this thing or that was your interpretation of their - ?
Hewish
I think the simplest thing to do is get the (?) Have you got these things around?
Sullivan
No, I don’t have that one – I do have a number of reprints here but I can check that.
Hewish
I think they probably said something to the effect that an irregular plasma would smear this thing out – something of that sort.
Sullivan
But you worked it out in much more detail - ?
Hewish
Well, I put numbers into it.
Sullivan
Now – well this is answered for me. In ’55 when the 2C survey was published, you and Ryle published a paper describing the aerial -
Hewish
Good God, I’d forgotten about that.
Sullivan
And in the Memoirs also.
Hewish
Oh yes.
Sullivan
But I didn’t realize that you were an aerial designer and I was just wondering how that came about also.
Hewish
No, no. The dipole arrays that went into that. I designed most of it.
Sullivan
So I should ask you about the 2C aerial. Were there any technical difficulties in the designing of that or was it - ?
Hewish
No, not really. I mean it was a straightforward antenna design.
Sullivan
Once you had the concept of what you wanted to do -
Hewish
Arrays and so forth. One had to design matching devices and so forth and get out there and do it. No, there was no real problem. It was just straightforward engineering, I think.
Sullivan
Next thing I have here is the Jodrell Bank Symposium in 1955 you talked about – I guess this was not actually your own work so much – solar corona with variable spacing interferometers – that was O’Brien and and… reporting (?)
Hewish
(?)
Sullivan
I’d be interested in – if you could tell me about that meeting. That was really the first meeting -
Hewish
First IAU Symposium on radio astronomy, wasn’t it?
Sullivan
Right. As to really what was the atmosphere – what was going on?
Hewish
Oh, it was quite interesting really because it was the first time the Russians had been let out, really as far as I can remember. (?) And I believe that was the meeting where they claimed to have discovered radio emission from deuterium.
Sullivan
I think that’s right. Or was that the Paris Symposium? I’m not sure which.
Hewish
I’m pretty sure it was at Manchester, actually. I think that happened before. And the meeting was hushed, as it were, while this announcement was made. And I don’t think people believed it. You may be right – it may have been Paris; I can’t quite remember now.
Sullivan
No, I think you’re right because – but I had the impression… At Jodrell they went and tried to check it and then they reported on that at the Paris Symposium.
Hewish
Hmm… But the interesting thing from my point of view I think was that I discovered that the Russians had been doing very parallel work on the – Vitkevich had been doing the same sort of work and we got in some arguments and debates about interpretations. And that – it was the Manchester Symposium where I first met – you know this guy V. V. Vitkevich?
Sullivan
Well he’s died now unfortunately.
Hewish
Oh, he’s died -
Sullivan
I would have liked to talk with him. But he was publishing work and you were unaware of it because you just - ?
Hewish
Well, we didn’t read Russian (laughter). We didn’t – you don’t cover all the literature. Science wasn’t the same then as it is now and I suppose that’s really bad for us but - No, I discovered then that he’d been doing similar sorts of things. But you know, there’s always been a bit of imitation, I mean, the sort of things they were doing always looked so closely parallel to ours that I was very suspicious about it actually. When they said, you know, we’ve been doing this a long time before, I was never quite sure about that actually. Because everything that was done looked so similar. However, -
Sullivan
Well maybe that’s the proof of the scientific method has some validity or something -
Hewish
Could be, yeah.
Sullivan
But what about the state of radio astronomy at that time? Was this really the first chance to have an overall worldwide assessment of it, would you say?
Hewish
It was more or less. Yeah, I think it was. This certainly was the first meeting I’d been to. There’d been an IAU meeting, I think, or was it URSI meetings that Ryle went to - ?
Sullivan
Well, ’52 I think -
Hewish
There was a ’52 meeting in Rome. That’s right. Well now Ryle went to that but we were – it wasn’t so much IAU as URSI it was the forum in those early days. It was the radio technique and it wasn’t really… We weren’t recognized as astronomers, I don’t think, in 1955 – scarcely at any rate. But - no, I recall it as being a sort of meeting where your program of work is changed by going to it. It was a real informative meeting, not like this thing that we’ve been having here, which doesn’t make any difference as to what you do. In those days – in the van coming away from it, I remember we decided on a complete program of work.
Sullivan
Can you remember specifically what - ?
Hewish
Yeah, more or less. This was – it was, I think, in fact, in the van we all drove up in this crummy old Hillman van which had punctures here and there – I’ve got a nice picture somewhere of Martin Ryle mending a puncture. We actually camped in the Welsh hills on the way to Dublin. That was when the IAU was in Dublin.
Sullivan
That was in ’55 also?
Hewish
That was just – I mean the Symposium just before that and we went straight from one to the other.
Sullivan
I see.
Hewish
And camped in the Welsh hills. No, it was, in fact, can remember driving along in that van between Manchester and Hollyhead where we were going to take the boat across and deciding what we really should be doing was improving – was getting more accurate positions for these radio sources.
End Tape 60B
Click start to listen to the audio for tape 61A of the 1976 interview.
Sullivan
This is continuing with Hewish on 19th August 76. Minkowski gave a talk there which might have interested to -
Hewish
Oh, indeed, yes. That was when – of course, the first – well the optical identifications were known before that, weren’t they?
Sullivan
‘51
Hewish
But that was the first time they came out and we really saw them. But no, I think – what I was about to say was I think it was then on that actual journey that we really decided what we should be building was a four-element interferometer. The thing from which 3C actually finally derived but it was actually not going to be a -
Sullivan
4C - ?
Hewish
No, no. I think 3C. I mean 2C… well, 3C hadn’t happened then -
Sullivan
No, 3C had not yet been done. It’s always been my understanding that there were always plans to do two frequencies with the same aerial.
Hewish
Two frequencies?
Sullivan
Yes.
Hewish
Ah, I’m not talking about two frequencies. To my recollection the surveys that had been going on up to then had been done with the extensions of the Long Michelson.
Sullivan
Right.
Hewish
Right? And that was essentially – well, I mean that was the 1C era. The 2C survey was made by John Shakeshaft and others with a four-element interferometer. Now that was the brainchild that we, more or less, dreamed up on that trip except we were not going to -
Sullivan
But now hold it – that was published in ’55, you see. The same year as – the Dublin.
Hewish
2C, but I’m not talking about 2C. I’m talking about what’s coming is 3C. Oh, just a minute. My memory is getting bad, isn’t it - ?
Sullivan
The Dublin thing was ’55, and 2C was published in ’55. Therefore, observed –
Hewish
2C was published in ’55?
Sullivan
Two or three years before then.
Hewish
Well then maybe - there must have been some other - I’m confused - There must have been some other meeting. We started to build an antenna and, in fact, never completed it, because the idea was to have an interferometer pair like this and then an interferometer pair like that.
Sullivan
Two diagonals?
Hewish
Two diagonals and compare those interference patterns to get accurate declination. But in the end we built all four. And the other one never got built. We started to (?) around, but never completed it. Of course, this other idea of having actual (?) overtook it. So that must have been – that was the first 2C thing. And so this discussion must have been -
Sullivan
What was the other idea?
Hewish
Just to use antennas in pairs, I think. A pair like that, and then a pair like that. You used crossed interferometers. It was, in fact, in the end as you know we connected them up like this, and did all four at the same time. It was a different… And this was to have been a much simpler device this finally turned out to be. Yes, you’re quite right that – 2C had been going for a long time.
Sullivan
Any other recollections from the Jodrell Symposium?
Hewish
That was also a recollection of a Russian theorist whose name escapes me, was quite impressive -
Sullivan
Talking about synchrotron -
Hewish
Talking about - no - Well, it may have been, yes. But it wasn’t the guy that invented synchrotron - so to speak. No. it’s a long way back. But I remember it as being a sort of lively meeting where people hopped up and down – it mattered what people said and you were annoyed when people contradicted you.
Sullivan
I’ve heard someone say that Martin Ryle got up with a long strip chart of 2C survey, I think it was this meeting. And someone else assisted and held it up on the stage and the audience was just – almost again – sort of overtaken, in awe of all these sources.
Hewish
In awe of all these sources - of the sources coming out. Yeah. That could well be because, yeah, that’s right, there were sources everywhere. That survey, that’s right – that rings a sort of bell.
Sullivan
It was still an exciting point at this meeting.
Hewish
Oh yes, surely, surely.
Sullivan
Next thing I have here in terms of publications is the Crab Solar Corona Occultations in June ’56 and then – I guess you were doing them virtually every year.
Hewish
Oh yeah, more or less.
Sullivan
Consistently improving - What were you finding out as you went through the ‘50s?
Hewish
Well, the thing was to extend it to as far away from the sun as possible. Try and get the corona (?) distribution. And then finally to measure the scattering in three different directions essentially, using triangulated interferometer system. To see if one could get scattering anisotropies which would tell you something about the filamentary structure of the corona. I mean, one had seen that all corona filaments close to the pole and the idea was that if these things extended further out you could see which direction the filaments were in. And nobody had any ideas then about the solar magnetic field. Any distance from the sun. But we all knew elementary plasma physics and knew that the field would influence the plasma and you’d get the frozen-in configuration. I remember looking particularly for a dipole field in the sense – mean everybody knew that the sun had a north and south pole and it was thought that the dipole field line would be going like that. So that if you had filaments lined up parallel to these things you ought to get scattering which was preferentially roughly at right angles to the magnetic field. And the idea was to look with triangulated interferometers to see if these things actually happened. And the first surprise was that the field lines never bent back, I mean they went out straight.
Sullivan
About this time there were intimations of a solar wind. I mean Biermann and others were talking about that comet tails might – Were you tied in with this?
Hewish
Yes. A bit. It was consistent and there was paper published before the solar wind really was a term in use. Where we suggested that the flow, in fact – that the field lines that we were getting suggested the flow – it might have been a flow like that. This was a paper I wrote - oh, it was a Monthly Notices paper -
Sullivan
Before 1960?
Hewish
1958 -
Sullivan
It may not have made this radio astronomy bibliography because it was deemed to be more straight physics or something -
Hewish
No, no, I’m pretty sure it was before 1960.
Sullivan
Well, I see. I have a list of the ones afterwards - Oh, here you are - Hewish and Wyndham ’63. “Solar Corona in Interplanetary Space.”
Hewish
That’s probably it.
Sullivan
That’s the one.
Hewish
That may well be it.
Sullivan
I just have a list of the ones after 1960. I didn’t have the abstract.
Hewish
We specifically mentioned our out-flow then, certainly.
Sullivan
Right, but by that time there were spacecraft observations… What about back in the mid-50s. I mean Biermann’s ideas came up in the early 50s.
Hewish
Well, I think the thing to do is try and find the first published paper with – I can’t remember so far back… I know I’ve got some papers somewhere about the fields – there was a paper I published here on (?) this sort of thing where you publish the actual direction which the scattering went. There were papers prior to this – there was a paper I published with Gorgolewsky.
Sullivan
That’s right. That was Observatory, 1960.
Hewish
Yeah.
Sullivan
Crab once again on another occultation.
Hewish
Yeah.
Sullivan
So you think it’s fair to say that this – these observations did not really contribute to the idea of a solar wind?
Hewish
Ah, oh, they certainly contributed to it I think, because there was evidence of density far from the sun. I mean, they couldn’t have been dropping off like (?) -6 or something (?) And it was, after all, the only way of looking at the solar plasma out there rather than very rare total eclipses. And Parker was very glad that we were getting stuff. He used our data as corroborative evidence that there was something a long way from the sun. But we weren’t talking about winds, I mean we were just trying to get the density variation of the thing.
Sullivan
Were you able to measure velocity in any sense?
Hewish
Oh, later, but not then.
Sullivan
But not in the 50’s?
Hewish
No, not in the 50’s. This came when we had this - the (?) scintillation came. But certainly not in the early days.
Sullivan
Anything else here before ’60 - That looks like it at least from your publications. You were not directly involved with the whole 2C controversy?
Hewish
No.
Sullivan
Well, let me ask then about – the next big thing really was the early 60s when you came up with interplanetary scintillations. Had you thought about this as a possibility before or was it a matter of -
Hewish
Oh yes, surely. The actual date of the notebook – I mention in the Nobel lecture. But the possibility of scintillation we thought about that years before. I’ll have to look back and see just what date it was. But is early in the occultation. When I was on the ionospheric scintillations stuff extending it to the solar corona. The immediate question was why is it just broadened – why don’t you see actual fluctuations in the signal. I mean, that’s what ionospheric scintillation is. You don’t see the source jiggling about in the sky – what do you – you see intensity fluctuations. I remember we did calculations to – and came up with the notion that these things, the fluctuations would be about 1 kHz in speed in (?) and but to see them at all you’d need a source with a ridiculously small angle size. – seconds of arc, or something.
Sullivan
Because the screen was so far away?
Hewish
No, because the, because the scattering was so great. No one had ever done an extrapolation and said, well, if you got 10 times further from the sun and there really are sources a 100 times smaller than we know about, then this thing could be detectable. No one ever thought of being such a mad fool as to look for it. And I’ve been kicking myself solidly ever since because it would have shown up quasars years before they happened if anybody had actually bothered to do that experiment.
Sullivan
And technically it could have been done in the early 50’s -
Hewish
Yes.
Sullivan
How was it that you come upon the interplanetary scintillation?
Hewish
That was quite straightforward. We’d been doing a survey of accurate positions looking at certain sources over and over again using – let’s see what antenna were we using then? Well, this would have been done at Lord’s Bridge so this would have been done on the 4C antenna but not using synthesis. This was work done by Margaret Clarke and others – were certain sources were being looked at, day after day to get the position more accurately for identification. And one or two of these sources showed peculiar patterns where the fringes were slightly unequal in size, and consistently on some sources. And Margaret Clarke in her thesis noted that these things were not so very far away from the sun. I mean, in the sense that within (?) or something like that. And the idea of scintillation came up in a discussion on Saturday morning – we always met on Saturday – we were all talking around these things. And somebody said something about ionospheric scintillation, you see, and it was round about then that Henry Palmer was getting the angular diameter of these sources and showing that they were less than a few seconds of arc or something. And someone said, “Well, why isn’t it like ionospheric scintillation?” And I went over the numbers again in my head and the pin dropped, splash. Well, it could happen. At first, well, I thought, well, it’s crazy, it can’t, because I’d already remembered the earlier argument, you see. But when you’re thinking about it again - And so we did specific experiments on (?) on 3C48 which was then one that was known to be of 2 seconds of arc or something. And, lo and behold, there it was. I mean, you know the thing was fluctuating. So that’s how it happened.
Sullivan
So what you’re saying is that these earlier calculations on the solar corona had sort of given you a mental block, more or less -
Hewish
Hmm - Well, I mean, you need a source a 100 times smaller than anything known. Yeah, that’s right. One had just forgotten about it – put it to one side.
Sullivan
I forgot to ask Martin Ryle about it, but I won’t forget to ask you. Namely, about the Sputnik - I’d heard that this – well this was on a Friday, October 4, and can you tell me what happened on that Saturday at one of those Saturday morning sessions?
Hewish
Well, we said something about why the devil don’t we build a simple interferometer to look at it. And I wasn’t actually involved with the building of it but a couple of guys went out (?) and somebody else and matched up a couple of dipoles and just strung them out on the ground virtually at the rifle range, which is not far from here and got the fringes. And then, of course, the balloon went up, I mean, this thing was in transit and we knew where it was pretty accurately. And Jodrell Bank was able to get onto it and crack it but until then hadn’t been able to pick it up because one didn’t know the position of the thing.
Sullivan
Were you part of the group that was working previously on this for a couple of weeks?
Hewish
Oh, yes. Well, everybody did, actually. I mean, there wasn’t much to do once it had happened. We went up there and listened to the thing going, “bleep, bleep, bleep.” But, I think everybody (?) then. I think I wasn’t so excited as some other people about it, I don’t think at the time, actually.
Sullivan
I’m cutting things off in general in my study in the early 60s so I won’t ask about further details of you ionospheric work, but there are a couple of exceptions I’m going to make. And one of those is certainly the pulsar business. I understand that this fellow (?) has interviewed you, is that correct?
Hewish
Yes.
Sullivan
For his things, so I won’t belabor the point because he’ll be published that. But, I still would be interested if you could tell me basically how this came about from your point of view?
Hewish
Well, we couldn’t have missed them. We were doing sky surveys – which part are you actually interested in? I mean, there’s no point in going right back to the (?) The point was we were – I was – interested in doing the sky survey on scintillating sources just to get angular sizes and built the (?) array to do it. And we just were plotting every source that showed a fluctuation wherever they occurred. And it wasn’t long before Jocelyn Bell, was working on the project, had caught them. I mean, the project was to analyze the records, continuous records, and just write down everything that fluctuated. It wasn’t a question of analyzing them to pick off the interesting things. We put down virtually every fluctuation because we knew that there was going to be a lot of interference – there was a lot of electrical interference. And, I think, personally, what showed them up fairly quickly was a chart we plotted where the idea was to put on a crude sky chart – this is something I asked Jocelyn to do – put crosses wherever a source happened – you know, in RA and dec. Now we were making repeated surveys every week of the sky that was looked at again for the scintillation survey. Of course, this kept on inking themselves in – they were genuine. If they’d just occurred randomly, then, you know, it was just sporadic interference. And when this one had happened about twice, what Jocelyn noticed was that – I mean she was analyzing the whole record – but normally you don’t get much going on during the hours of darkness when you get ionospheric scintillation which causes these things. But when there isn’t ionospheric scintillation, you don’t get much interplanetary scintillation because it’s round the back side of the sun. (?) so the direction where it drops to a pretty low (?) And she brought it to me and said, “Look, isn’t it funny that we’re getting this interplanetary scintillation round the back.”
Sullivan
Even at night. Was the idea of doing it every week to new sources (?)
Hewish
No, it was a technique. You want to watch sources (?) that’s the way we actually measured (? Entire phrase not understandable)
Sullivan
Paul Scott made an interesting comment and I’d be interested if you agree or not. That the 4 original pulsars were sort of the only ones that could be found. They had the right dispersion intensity and so forth.
Hewish
That’s pretty much so, actually. We found some more later. The number went up to about 6 finally.
Sullivan
But it was rather discouraging that you were expecting really to come up with quite a bit more.
Hewish
Well, not terribly really. I wasn’t very discouraged, actually. For a long time, we had 6 and that score wasn’t beaten for a year to two. When the big telescope turned on it, they didn’t turn up pulsars very quickly.
Sullivan
Is that right?
Hewish
Yeah, no they didn’t. If you look back in the literature, I think our 6 … I mean, at the end of the first two years there was something like 20 or something. And we discovered more than a third of the whole shoot. Arecibo didn’t get them for a long time. Jodrell Bank got a few. They got a few. But the numbers were very low the first few years. We realized this dispersion business but the antenna wasn’t really well suited for pulsar work. It was very well suited to make the initial discovery. It couldn’t have failed to have discovered pulsars. However, you don’t like the records. Whoever was on the job. It would have been gross negligence not to have seen them because they were the biggest scintillating sources in the sky sometimes. We were on to them quickly because the record analysis had been done pretty smartly. And that was good. I’m pleased that we did, but if we hadn’t gotten them, my God, we shouldn’t have been doing radio astronomy. It was just great good fortune to be in that era.
Sullivan
And what happened is that once you had these strongly scintillating sources and began looking at them and then it became pretty clear that there was a periodicity in it.
Hewish
Well, it wasn’t clear for a long time, no. (laughter) Well, you can read all this history – it’s all written down. We didn’t believe it was a genuine signal till 6 months after we were seeing it. And getting pulsars made us disbelieve it even more.
Sullivan
You didn’t believe the scintillating signal either?
Hewish
No, no. We started picking these things up, you see, and initially we thought, “Well, it could be a source of very small diameter scintillating very strongly around the (?) solar direction.” But since the record looked a bit funny and was just really incidentally the transit time of this thing can shift by up to one or two minutes, we thought it was somebody mucking about. We didn’t think it was a proper signal until it had reappeared about 6 times. The first time it ever appeared on a record was the middle of July and that record had been analyzed in August sometime. So it was, in fact, September-ish when we began to get really suspicious about this thing.
Sullivan
Why did the time of transit change by a minute or two?
Hewish
Why, because the source switches on or off and we got a broad beam and so you sometimes see it -
Sullivan
Oh, I see. So it really is more time detection which you were taking than the time of transit or something like that.
Hewish
Well, it was confusing. And it was for these reasons it looked wrong. We knew what scintillating sources looked like and this looked a bit funny because the beam was too narrow sometimes. And, as I was saying, it shifted because of that, you see. So we set up – we were doing experiments. I was concerned with scintillation theory and getting as rapid a recording as possible. We got a special system set up in September using a new fast recorder for looking at 3C273 which makes a close transit across the sun. And I wanted to test out the scintillation theory. And we had this thing set up and I asked Jocelyn to put it on so that we actually looked at this strange thing as well to see what character the signal actually had. Because if it was interference they wouldn’t look like scintillation. They’d look like something else. And nothing happened till November 28 because – well the special experiment had priority on 3C273 and when that was finished, that was coming on to October. Jocelyn then ran the thing for several days looking at CP1919 as it was later called. And nothing happened, she didn’t get anything. And she was getting a bit fed up when all of a sudden it came back. And that’s when she got the first pulses.
Sullivan
So it didn’t really have that high a priority during those first few months.
Hewish
No. There wasn’t much to do about it. If it was real, it would keep repeating itself. I thought it was a flare star actually in September because we could just about detect it. Sounds like the sun, a bit more violent. We could have detected it with the telescope and solar activity we all knew about. It looked a bit like that, a fluctuating signal. So I thought probably we’d detected that and some of those calculating actually came out in the first publication. That if it was a flare type, you know, something like a solar flare it could have been detectable at the distances of nearby stars.
Sullivan
Well, one final thing that I’m interested in is the whole business of, you got this completely different sort of signal that’s ever been gotten in radio astronomy before. And you’re obviously thinking now, “What do we do with this?” and thinking of the possibility that it was an intelligence signal and so forth. And someone has mentioned to me that there was real concern for a few weeks, well, shall we notify the government, and so forth.
Hewish
We didn’t know what it was. If it could have been intelligence - I mean, you can’t forget about the possibility of picking up intelligence. And this seemed to be the most likely origin for a while. The moment the accurate timing of the signals was known, then I realized we could do a Doppler shift experiment to actually get all the motion, which would have been a good test for that. But talking about it with Martin Ryle, we said, “You know, if it turns out that the intelligence is the simplest explanation, who do we do with it?” We never really answered that question. But you can’t just publish it as science and it isn’t science. We were going to have some sort of a – well it wasn’t really a Press Conference – I mean, we were going to go down and talk to the Royal Society. That was really what was in our minds, that we would have people like Blackett and scientific big guns and put it to them, what do we do about this. But, keep it amongst top scientists to begin with.
Sullivan
Yea, sort of black cloud -
Hewish
Yeah, in a sense. We didn’t really quite know how to handle it. And we felt we had to talk to somebody.
Sullivan
Why wasn’t it a science? Seems to me like it would be science. It would be, perhaps, the greatest scientific discovery of all time.
Hewish
Well, no, it isn’t. It’s a message. It isn’t astrophysics if somebody is signaling. What I’m saying is, you can’t just publish in Nature saying, “I have discovered intelligent signal from outer space.”
Sullivan
Well, that’s an interesting philosophical point, but I hadn’t thought about that.
Hewish
Well, it isn’t. They’re not natural signals.
Sullivan
Right. Would you argue that all the experiments that are beginning to be done now to look for extra-terrestrial intelligence are not really science?
Hewish
Yeah.
Sullivan
It’s some other activity of society?
Hewish
Well, it’s scientific in a sense, but it’s not what I call science, no.
Sullivan
I’ve never thought of that. I’ll have to think about that some. When you were thinking that it might be intelligence looking for various sorts of modulation – is this right?
Hewish
Yeah, that’s right.
Sullivan
And what sort of approaches did you take?
Hewish
Oh, just common sense. I mean, really once you 3 of the pulses to look at you look at them and say, “Are they telling me anything?” There wasn’t much sophisticated you could do because we couldn’t break the pulse down. The sensitivity was only just enough to pick up the pulses.
Sullivan
Oh, I see. You couldn’t break the pulse down. That’s what I was thinking about -
Hewish
You get a chart running all the little blips like that on it, most of the time. And you know you’ve just detected these things in the noise.
Sullivan
So, really the amplitudes of the pulses and -
Hewish
Yeah, that’s right. You looked at them and said, well you know is it going 3.3.1 or something like that. It wasn’t doing anything logical. We did cross correlation analyses on these records which would have shown up any pattern.
Sullivan
For this reason, yeah. Well then what was it that finally took you away from the idea of intelligence?
Hewish
Two things: first of all, the Doppler Shift showed there wasn’t any.
Sullivan
How long did it take you to establish that?
Hewish
We began timing them on December 11 and it was just about New Year when we got enough data to actually plot these things out properly and show that there was a Doppler shift – the big Doppler shift – but it was all (?) motion (?) When you corrected for that there wasn’t any residual which could have been something else. I reckon it took about two weeks to really know that answer. During that time, we’d come up with one or two more pulsars actually. You know, suspicious looking things which looked as though they were pulsars. These were confirmed early in the new year and I don’t know why it should be, but if you’ve got two or three it’s less likely to be intelligence than if you’ve got one. I don’t know the logic of that but -
Sullivan
Yeah, I’ve thought of that myself.
Hewish
To my mind, it was lack of any Doppler shift and the fact that this thing – if it was in the (?) orbit, you know it was in such a long period orbit that it wouldn’t have been more than enough to support life. It couldn’t have been near enough to any sensible star.
Sullivan
Yeah, or any sort of life that we’re – once again black clouds would not really be ruled out. Did that bother you at all?
Hewish
No. (laughter) In fact, by about early in the year we were quite happy that this was some kind of star, you know, we were looking up at white dwarfs (?)
Sullivan
Was the discovery of the other pulsars a deliberate search to see if there are more of these things?
Hewish
Oh sure. When we knew what to look for – well Jocelyn then went back through all the records. 0950 I remember, I’d spotted that one sometime earlier on while we were wondering about CP1919 that this thing – I’d seen a source on the record that looked awfully suspicious. That turned out to be 0950. And Jocelyn found 1133 and - Anyway, we looked up possible pulsars and did searches night after night, checking off and some were and some weren’t.
Sullivan
Ok, to close, do you have any comments about how you’ve seen radio astronomy grow over 25-30 years almost?
Hewish
Oh well, it’s settled down now to a steady bond with astronomy, I think. What I’ve seen is the heat go out of it, I think. I mean, the 60s were sort of Golden Age when everything was happening.
Sullivan
You think even more so than the 40s or 50s?
Hewish
Oh yes. Well, the 40s and 50s were enormous fun. No, no, they were tremendous. I mean, when the first redshifts came in, the discovery of Cygnus and so forth. That was a marvelous period. But we were still talking about colliding galaxies then – right up almost to the 1960s, people were talking about colliding galaxies. And we really didn’t know what was going on. I don’t think we were proper astrophysicists in any sense then. But the 60s, I think, were the Golden Age when the quasars really raised their heads. Molecular lines came after that but then there was this business of background radiation and pulsars. And those – one couldn’t really have kept that sort of excitement up. And I think, the late 60s and merging into the 70s and now almost into the 80s things sort of leveled off. It’s steady systematic work on all these different things that’s going on. And when you come to a conference, people aren’t jumping up and trying to turn the world up-side-down like they were in the early days.
Sullivan
Oh, yeah. Well, I’ve heard people – most of the people that were working in the late 40s say that that was an extremely exciting time.
Hewish
Oh, yes, it was.
Sullivan
Every time you build a new aerial you discover -
Hewish
Sure, that’s right, yes.
Sullivan
But are you arguing that the 40s were that way and the 50s not quite so much and the 60s really took off?
Hewish
There were sort of two phases. First of all, there was the wild surmise, you know, that the sky was full of these peculiar objects which were at large redshifts and never knew what they were. I mean, Cygnus type objects. But the phase then was to build bigger telescopes, get better positions and just really count them. There was all the excitement about cosmology and the first source counts. That was terribly exciting when the first source counts showed this wild divergence.
Sullivan
You said a wild surmise about the fact that most of these sources were extragalactic – what do you mean?
Hewish
Oh no, I’m sorry. No, I mean the first telescopes were showing radio sources everywhere and nobody knew what they were. And there were just one of a handful of identifications. The push in radio astronomy was to make bigger telescopes, more sensitive telescopes and get many more of them. To get better positions. There was no doubt what anybody ought to be doing. And it was just simply that.
Sullivan
I see. And it was a bit more of a drudgery sort of thing are you saying -
Hewish
When?
Sullivan
In the 50s, I mean, you just had to continue doing the same thing only try to do it better.
Hewish
No. The 50s were sort of in a sense consolidation – really establishing what some of these things were – these radio sources, careful position measurements that led to the discovery of Cygnus. And then the first systematic survey is the cosmological possibilities. The cosmological things were of course a tremendous excitement when we realized that source counts could tell you something about cosmology. But there was then the phase when it was realized that you had to have really rather more sophisticated techniques and synthesis came along and that took a time to get operational. But then with the synthesis on the one hand showing you all these wonderful double objects and the long baseline interferometry showing you some of these highly compact objects. The first discoveries of the first quasars and then the other things like the microwave background and pulsars and so on. That was a sort of second great boost of excitement. I mean, really it was exciting all the time.
Sullivan
But I see what you’re saying.
Hewish
But what I’m saying, it sort of goes in waves. I’m looking back to the mid-50s – it was all tremendously exciting then. But on a much more – in a sense – amateurish level.
(Interruption here)
Hewish
So it’s settled down now and people are going a steady systematic spade work. And you can’t expect it to be like it was.
Sullivan
But are you sort of waiting that any year now – it’s sort of inevitable that – it seems to be for instance the resolution of what’s going on in the quasar – for instance, maybe it will become established that there are non-gravitational redshifts - something like this could really bust open physics.
Hewish
Yeah. I don’t expect that. I take a fairly conventional view. It seems to me that (?) Ok, there are high energy sources, but there’s a lot of galaxies now that require high energy and quasars aren’t special in that respect – the infrared galaxies and so forth. No, I think physics is going to be all right. I doubt if there are black holes. Even if there were, it wouldn’t bust open physics, I don’t think. I mean, you understand physics and black holes – it’s getting along quite nicely, isn’t it? But, we’re not going to – the instrument has, more or less, reached the end. I think there are not many avenues of pure investigation that are going to show up wildly new phenomenon. We know how big the quasar sources actually are. We know something about the motions of them, these highly compact objects. Things like the VLA are not going to show up things we don’t know about.
Sullivan
Well, that’s a very controversial statement.
Hewish
No, no. Why would you say that? I mean, one knows (?) angular structure, more or less, (?) and one knows what resolution (?) and I can’t see where they’re going to turn up many funny things. It isn’t a highly sensitive instrument, it’s an instrument for high resolution. And one knows now, more or less, what structure is left to be resolved. There isn’t very much. It will have a comfortable resolution to the point 3 arc seconds you can push ours to 15 GHz. There’s not really going to be an awful lot -
Sullivan
You don’t think for instance many detailed maps of the very inner nuclei of galaxies might resolve the problem of what powers (?)
Hewish
Oh, oh yes. But what I’m saying is we know roughly how big these objects are already and we’re producing maps of them just like the early radio maps. And showing that they’re little complicated things. The rate at which one can actually produce information of this sort is so slow that I don’t see any radical changes coming quickly.
Sullivan
Now when you say radical changes, do you mean in that no qualitatively new phenomena will be found?
Hewish
Yeah.
Sullivan
As opposed to astrophysical understanding?
Hewish
Yes. Clearly what we need now is decent high resolution maps of the compact objects which requires synthesis on a global scale. Now that’s going to be awfully long and a hard problem to do. And already we know roughly what these things are. There are bits flying out from the center and you can see something of what’s going on in there. But my guess is that it will be more than 10 years before we really know an awfully lot more about this. And then what use, you could – what use is that without the optical data to go with it? The resolution of this far exceeds any optical data to go with it? The resolution of this far exceeds any optical picture now. And radio on its own is a bit of a weak tool. It’s when you combine it with optical spectroscopy and all the rest of it that you really begin to find out what’s going on.
Sullivan
Well, in fact, I was going to suggest that you might say that it’s something like the LST, would really be more likely to come up with some real fundamental discoveries.
Hewish
I would say so, yes. I think to achieve radio optical resolutions (?)
Sullivan
1/100 of an arc second with the optical would be an entirely new thing.
Hewish
Yeah. I think we’ve got to wait for the optical astronomy to catch up in techniques, and produce maps of resolution. Yeah, I think that’s probably right. But there aren’t many bits of the spectrum unexplored. I mean there’s millimeter, maybe some new phenomenon in millimeter.
Sullivan
But for instance, the whole infrared spectrum has not been looked at in detail. Infrared spectroscopy for instance could bust open -
Hewish
Oh, any new spectrum obviously can have surprises.
Sullivan
Well, thank you very much. That ends the interview with Hewish on 19th August 76.
End Tape 61A