Oral History Transcript — Dr. Emil Wolf
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
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. Disclaimer: This transcript was scanned from
a typescript, introducing occasional spelling errors. The original typescript
Access form | Project support | How to cite | Print this page
See the catalog record for this interview and search for other interviews in our collection
Emil Wolf; September 23, 1984
ABSTRACT: Comments on some of Wolf's research: first studies of coherence in the mid-1950s; lasers and the problem of complete coherence; holography; global coherence; the structure of laser light in the focal plane of lenses; coherence in the space-frequency domain; the nature of scientific progress. Also includes discussion of the social circumstances of his research: funding arrangements; consultancies; scientific contacts; the controversy between Glauber and Wolf, Mandel, and Sudarshan.
Bromberg: Why don't we start with what the environment was like at Rochester as you came here in 1959? Why don't we start there, with coming in and , see, you were not at Edinburgh, you were at Manchester ?
Wolf: I was at Manchester at this point. Yes. And Robert E. Hopkins came to England towards the end of, I think it was 1958 or thereabouts, specifically to Manchester to, to talk to — he was at some conference in England and he came specifically to Manchester to discuss with me the possibility of my joining the Institute. We exchanged one or two letters before, and he spent the day with me at Manchester, and we talked about it, and as a result of this and some correspondence which then followed, I came to the United States in I think it was April, ‘59, with an appointment , at first a visiting appointment, at the Institute of Optics.
Hopkins actually told me in England that the future of the Institute is a little bit unsettled. Up to that time it was an independent part of the university, reporting back to the president. It was not part of the college structure. But there were some problems with it apparently and I still remember him saying that he hopes that if I come to Rochester, and start some new project and so on, that maybe that will preserve still the independence of the Institute, rather than attempt to make it part of the colleges. Actually a few months later, it did become part of the College of Engineering, and that's when myself and several other people were offered to move to the physics department, and I should add that I had a joint appointment in physics shortly after I came here. So did several other people, and most of us did in fact accept positions in the physics department, but we had joint appointments in the Institute of Optics for a while.
Now, as far as research is concerned, in the Institute, and my own other activities, I think Hopkins, whom you'll speak to later today, is in a better position to comment on it. I do not have a recollection that much was done in the way of laser physics, although soon after the first laser became operational, there was an effort to do laser work, and in fact there were people hired specifically for that purpose. One person I particularly remember is Carroll Alley, who is now on the faculty of the University of Maryland, and there were some other people working with him there, Mike Hercher did very important work in the early days in laser physics, but as I say, Hopkins will probably remember much better, who the people were.
I don't know whether you want me to follow some of the other things you've outlined?
Bromberg: What was your interest in coming to the Institute is I guess the first thing I'd like to know. I'm going to move around. Why did the Institute of Optics interest you? Why Rochester, why did Rochester interest you?
Wolf: Well, the first thing of course is that my specialty was optics. It had been for many years, since I did my graduate work. And in fact, in the field of optics Rochester was probably the biggest center of optics in the whole world. There were at that time three or four other institutes, but nothing in the United States. There was one in Paris, one in Stockholm, and I think there was one in Italy, I think in Florence at that time, and that's about it. Of course my feeling about it was probably colored by the rest of the world (inaudible). So that was number one, but the other reason was economic. I found it very difficult to find a university post in England. In fact, at that time I was on a fellowship at the University of Manchester, and when you ask later on in some of your questions, you put down a few things connected with that, I want to tell you an amusing story. Somebody was sent to the United States to work with me, and by the time he came, I was at I was on fellowships for years.
That was a reasonably — it was a temporary appointment, with the understanding that if it works both ways, I would stay here permanently. So that's why I came to the United States, a combination of things.
Bromberg: You say that Hersher and Alley came later?
Wolf: I'm not sure about Hercher, but Alley almost certainly came after I think the first laser became available... You know, Alley worked with Dickey, he did his —
Bromberg: — yes, I know —
Wolf: I am pretty sure he came after me but you'll have to check — Hercher, I don't know: I think he was a graduate from — he might have been here already, he was, I think he was a graduate from but he was very early, in the early laser days. He got right into the field and was building lasers and doing interesting experiments and was actually quite a student. He left a few years later.
Then who were the big names here at that point when you were coming in, in '58 and '59.
Wolf: In Rochester?
Bromberg: Here, yes.
Wolf: Well, the names I knew in particular of course were Hopkins and Kingslake. They're in very different fields but their names were very well known to me. Kingslake, both of them, as you know, were leading designers of optic systems, both of them. So these were the main names, to me.
Bromberg: OK, so you came here and how was your work organized? Were you teaching a lot from the start or were you mostly?
Wolf: I was teaching a regular course, like the one on physical optics. I came in April, so it was a time you know when I couldn't start teaching anyway. I started in September.
Bromberg: I see.
Wolf: And then I started teaching a regular physical optics course.
So April '59 is when you actually —
Wolf: — Yes, that was accidental. There was some muddle about my visa. You know, I needed an immigration visa.
Bromberg: That's right.
Wolf: And Hopkins worked on the visa, and then got it and it turned out to be the wrong visa for me to be able to work here and it delayed my coming. I would have come at the beginning of the semester, the previous semester, but I didn't make it because of the visa. The whole basic difficulty was just the visa, is all.
Bromberg: Very frequently I'm getting in my interviews here one or another problem in connection with the McCarthy Era, by the way, but I'm assuming —
Wolf: — that was not the difficulty —
Bromberg: — from Bloembergen and so on, all these —
Wolf: No, there was nothing in it. It was genuinely not an immigration visa but a visitor's visa, and it wasn't the sort of thing on which they could pay me, but it was arranged.
Bromberg: Then another question I asked you was whether you had a specific research program, as you came here, to begin to see whether —
— I did — but you know, I noticed from your questions, you have very precise questions about this. The research program, I think, at least, it differs with different people, but my research program is not that definite, that I would say, this is the area I will continue working in. I was particularly , this was partial — it was well before the laser came on the scene, I mentioned this previously to you, and I was very much interested in problems connected with diffraction (increments? instruments?)
And this was the general area I was working in, and I was well aware that there were many unsolved problems, and I would say even now there are a lot of questions not settled. We will come to that later on some other questions. And I felt that this is an area in which I will be able to work for many years and find an interesting challenge in it.
You see, my first work, on coherence, started, I wrote that around 1953, '54. I came in '59, and since then I have been publishing on the average just on coherence, at least two or three papers a year just on coherence When I say coherence, I mean the whole general area of statistical and partial coherence, of course. Also I was publishing in other areas, particularly information in diffraction, based on and so on, but classical coherence has kept me busy all these years. It still continues.
Bromberg: So then I wanted to see if I could get some feeling for how being at Rochester and being in the United States interacted with your interests, you know, to set scientific problems.
Wolf: Well, that was really very much interconnected with what happened in the department of physics. Eventually I got a part time appointment, a general appointment in physics, and that was, I think, I would have to look up records, if you ever need to know it I could look this up, but I think I got an offer to join the physics department even before the Institute of Optics moved to the College of Engineering. I only became a full time member in the physics department afterwards. But it was clear, (Robert) Marshak at that time had the physics department, and it was clear that he was — I would like to correct this, it was first Kaplan and then Marshak. Kaplan I think was deputizing for Marshak while Marshak was in Europe, so the negotiations originally were with Kaplan. But it was quite clear to me fairly early on when I joined the physics department that they were interested in building up an optics group also. We called it optical physics in those days, later we renamed it optics. And it was given to me as a job to build it up. And my first choice of course was L Mandel, whom I knew in England very well.
And Mandel came, and then we hired Joe Eberly first as a post-doc to work with me. He came after he got his PhD I think directly from, or PhD, I think the negotiations with so that was the nucleus of the optics group in the physics department, Mandel, and then myself. The order I don't remember anymore. Mandel came I think three years after me, maybe four, we can check these days, but that was the nucleus of the optics group and that roughly continued like this, except everybody is moving up and all three of us are now full professors, but at one time we had a very large group, and we had more money than there is now. We had a group of about 20, 25 people, research associates, students — if you come to my office, I'll show you. I have a big picture in my office of the people, but that includes students of course.
Bromberg: That's enormous, though.
Wolf: Yes, it was enormous. Now it's much smaller. But not all that much smaller, because Eberly has got a large number of graduate students, and some post-docs, and Dr. Mandel has got also quite a few. Graduate students. So although it's not as large as it used to be, it's still a very sizeable group.
So then you ask about how the thing developed. It started developing around the physics department. I had appointments in the Institute of Optics on and off all these years. When I say on and off, the administration was changing, the policy was changing, — there was always some interest that I should be there, but at some point I was not keen on having an appointment simply because I was the only one in the physics department who had it and it created some problems because Mandel was not a member of the Institute and I was, you see. And that has been fluctuating on and off. I'm now being, I have now been at the Institute for the last four or five years, but it was always the relationship slightly unstable between the physics department and the Optics Institute, not for any obvious reasons except possibly a certain amount of competition amongst the administrative structures, where should the money go from the government, you know, should it go to optics in the physics department, should it go to optics in the Institute, and that unfortunately has slightly colored a few things in this.
But on the scientific level and cooperation, we have never had any problem.
Bromberg: Now, Mandel came here and he was very much interested in the relation between photoelectric statistics, so what was Eberly doing?
Wolf: What was Eberly doing? Mainly neoclassical theory, starting with James's work, developing James's work. James called his theory the neoclassical theory.
Bromberg: OK, I guess I don't know James's theory.
Wolf: Yes, it is another of these sorts of theories which broadly put are categorized as semi classical theory.
It was a theory which is not fully quantized — part of it was classical, partly quantized. And that was very popular for many years. In fact, there were several very interesting discussions which when on at Rochester Conferences. There's a history of this at every Rochester Conference until the last one. There were a lot of papers/on it, and there was this famous bet between — you know that.
Bromberg: Yes, well, it's in one of the things you say, one of your reports.
Wolf: Yes. But by and large now it's pretty settled that James's theory — it can't go very far, it's not a theory which answers all the questions, one cannot ask it quantum electrodynamics.
Bromberg: So the whole thing focusses on this border between classical and quantum.
Wolf: Oh yes —
Bromberg: — optics, as you begin to build up this group.
Wolf: That's right, and that was very much Mandel's specialty. When you say he was in the forefront of experiments and so on, you know, I knew Mandel in England. In fact, I brought him here because I was a great admirer of his work on, the early papers on they were quite outstanding, you know, the papers in 1956, and so I was very anxious to have him here, and he was here I think on two visits. He had appointments in summers and so on or a semester, before I persuaded him to come here permanently and he has been very consistent in this sort of work. He really developed the photographic methods for studying statistical properties of light to a degree which nobody else could have done, I don't think and it is still a very very powerful tool and is continually giving more results. At least, the most recent work, on the statistics, for example, it's first class work, and much, the and so on. He really pioneered all these things which started from his photoelectric effect.
Bromberg: Then did the laser, now, the laser is coming out on, it's being talked about as you come here. You come here in April ‘59 and you’ve already talked to Hopkins about it.
Bromberg: And everybody knows that people are working on it pretty hard.
Bromberg: Hoe does it begin to affect your own work? What did you start working on what were your first papers when you came here?
Maybe I should check on this .Quite an interesting question. Can you give me just a second so I see what I've got? I think in '59, the papers here relevant papers would be "60, "61 and so on, — sure, well, there were many. One of the papers is on correlation between photons and in coherent light. Well, it was the interest in correlation. This continuing interest in correlation, but then the paper of '42 which you I think wanted to ask questions about came directly as a result of the laser, some problems of coherent light. There are serious problems in understanding what true coherence is, and I must tell you there are still some problems unsettled to this day.
There have been many papers and it's far from settled, but I'm looking at this publication , my original publication around this time, obviously I was studying more towards understanding what full coherence meant, and broadening the basis of coherence, particularly things like the relation between coherence style, structurally, and so on, but of course, it was very directly stimulated by the laser, because as you know, one of the papers we published with Mandel, number 49 on this list, we show that one must be very careful — to define coherence style for something, spectrum which has many peaks, if you have many modes there, then the question of defining becomes very tricky.
But I should mention that this is actually my first paper on the subject of coherence. There's an earlier one but it was not oriented towards lasers of course, my paper number 32, published 1958, after I had been asking questions which turn out , could be broadened into the laser field and became relevant to it.
But the earlier work on partial coherence, as I mentioned before, I had no idea of the laser at all. But then it became, after I came to the Institute, slightly more directed to some questions of lasers, but the laser was never my real specialty. It had, my work had relation to it, but I would not consider myself as an expert on lasers.
Bromberg: So in other words, let me refer to this to see if I'm getting it straight, so it really is that you would do a piece of work and you'd say to the community, you'd call it to the attention of the community , to the fact that this had some interesting aspects from the point of view of lasers. I mean in reading —
Wolf: — it has bearing on lasers.
Bromberg: Yes. So that's the way we should read your papers in that case?
Wolf: I think so.
Bromberg: Not that the thing is inspired by lasers, or that the attempt to —
Bromberg: I want to ask about funding. When you came on, what kind of funding situation existed? Were you expecting to immediately get a grant or?
Yes, but it was made very easy for me, because Hopkins prepared the ground. He had good contacts with the Office of Scientific Research in Washington, and they were willing to provide money straightaway for organizing the conference. In fact, that was one of the first things Hopkins wanted me to do, when he talked to me in England. As soon as I came here he wanted me to organize the conference. He had wanted me to organize it from England, to get started, and I felt, I don't know enough about the American scene to know how to do these things so we postpone it. He would have liked a conference in ‘59 when I came here. I preferred to have it a year later, so it was in '60. Curious timing, you know — within two weeks after the conference ended, the first laser was announced, and there were rumors already at the conference here, about the laser action having been observed.
So as we got the funding, then there was no problem getting money for the conference, and the same people who gave us the money for the conference also provided funding for me for many years continuously. It started I think three and continued for many years from the Office of Scientific Research.
Bromberg: Was this a matter of negotiating what you were going to do, or did they just say "Whatever you want to do —''
Wolf: No, there was a genuine proposal, and by and large I must say they've been very broadminded in not modifying practically anything which I propose. It was always in the general area of diffraction and partial coherence and it continued. I never had any difficulties about the subject. But it wasn't just that they would give me a flat sum of money. There was always a proposal.
Bromberg: I see. But they didn't come to you and say "We'd like you to do a little more of this or a little less of that”?
Bromberg: And that was mostly for graduate students? What does one — or for freeing up time from teaching?
Wolf: What the money was for?
Wolf: Well, for several things. One is of course it supported graduate students, as you said. One is for research associates. Another is for publication charges. Travel expenses, secretarial assistance. You know, I don't know how it's done at universities but by and large it's almost impossible to function at the University of Rochester without outside support. I had, for the first time last year, I was without outside support, mainly because I was on sabbatical for a year, and the grant I applied for just before I went to sabbatical didn't go through and I had six months or so here without support. It was a terrible time. (crosstalk ) Even xeroxing, I have to discuss with my chairman to make sure I won't have any problems when I go to the xerox machine, no charges for them. Postage, long distance telephone calls, real headaches. I shouldn't say that I couldn't do it, but it always was a problem. I'd have to discuss it, I’d have to discuss it with other administrators and then the next round comes, you know, some of the things they gave me was an advance and so on. It's very strange to operate in a private university.
I'll give a little example of this which may amuse you. There is an Optical Society meeting in San Diego in October, and a short course is being given, and it's only three hours but still you have to prepare notes for students and so I've never given a short course, mainly because it's a lot of work but I'm giving one now, and the reason is simply that a year ago, when I knew they would be interested to organize short courses, I didn't have my NSF grant and I didn't know whether I would have travel money to go to San Diego. I have not missed a single Optical Society meeting in the United States, but it looked very far away, if I would have to pay for myself to go, so I decided to give this course because the Optical Society pays travel expenses and some extras, so that actually gives one an indication of how you have to operate.
Bromberg: That's fascinating to me. Probably I should have known that, but I did not know that that was —
Wolf: I wish I'd never promised this course. I mean, it would be fun to do it, but I'm, you know, so busy, and to find time for this, I will have to produce something like 100 pages of manuscript or at least of transparencies and so on, and bound copies and so on, it's quite a job, but I just was determined to go to the meeting and I didn't see any other way. In fairness to the university, had I not done this, there was a chance the university must have paid-occasionally the university pays one trip a year. But it's not enough, it depends on how many other people, there's a certain amount of money, say $10,000 a year for the whole department for travel. It depends on how many other obligations there are.
So a year or two years ago I was invited to give a talk at the Maxwell Conference in Edinburgh, celebration of 100th anniversary of his birth, and there was no obvious way to charge this to the university, but the university actually paid for it,, half of it. Half was paid by the organizers. But these things are not guaranteed beforehand, you know, it's not obvious that it can always be done. So this is just an illustration of it. And even now it's not the easiest, and NSF has been quite generous but I didn't get as much money as I applied for, and we have for example a budget of a thousand dollars a year for publication charges. Now, I can tell you, that doesn't cover even a quarter of the publication charges.
So I don't know what you can do out of it now and there's a possibility I may get another grant from the Air Force base, but even that won't be enough for the amount of publishing we do, so we may have to send in papers without paying for publication charges, you know what happens, they'll just delay publication of the papers.
Bromberg: No, I didn't know that. I know they don't give you reprints.
Wolf: No, the reprints they don't give any more. But it's sort of a slight pressure on authors to provide money. If there's no money, the paper is delayed. So it will be delayed. It's not a crisis. I'm not one of these people who feels he has to see something published the next day. But it's a bit of a nuisance. So the money isn't that plentiful, really. A lot of it is chance; you know, and so on, your contacts, whom you know. I've been very lucky over the 25 years I have been here. But there are occasionally little problems.
Bromberg: I think that's very interesting indeed. One of the questions I asked you, I got the impression from listening to some of the talk at the Symposium this past week, the Du Bridge Symposium that you organized, that there are just a lot of contact with firms like Eastman Kodak and at one point early on I asked you if in your own work, whether, when you first came, whether there was an important contact with Kodak.
Wolf: Yes, two companies in particular. Later I list here in my notes other companies I consulted for, but the contact with Eastman Kodak and with Batelle Memorial Institute in Columbus was extremely profitable scientifically.
Kodak was actually very generous to me. They offered me consults almost as soon as I came here and I continued for about 10 or 15 years, and moreover, they left me almost scientifically, anything I wanted to do there, and I teamed up with Eric Marchant, who is now retired from Kodak, but is now on the faculty here at the — as a part time professor or something, I'm not sure.
And we developed, we published quite a number of papers in coherence theory, and we developed an area which was actually originated by Adrian Walthers on what is known as radiometry and coherence, the foundations of radiometry. It's a very practical subject, to discuss; it’s concerned with the distribution essentially of energy from radiating sources. That's a problem which as time went on became more and more with lasers and there are still very basic problems there. Essentially, how the coherence of a source affects the angular distribution of radiation. I will have more to say about it later. It's a very basic problem, very poorly understood, and we have written quite a number of papers about it, Marchant and I. One of the papers was recently reprinted complete with the whole paper in a book by Marathy, a book on coherence. The whole paper is reprinted there.
So that was partly because, we were able to do this because Kodak essentially let us do within reason what we thought was sensible. I worked in their research laboratory, and Marchant was probably closest to me, in his research interests. He was a mathematician by training, with strong theoretical leanings.
So that was one of the very useful things which came out of my contacts with Eastman.
Bromberg: So it wasn't just going there and talking to them about their work. You were actually doing —
Wolf: — no, the Kodak people would come in and ask me questions about their problems here. But basically I would say 90 percent of the time; I was left to do what I liked pretty much.
The other contact of a similar type was Batelle Memorial Institute, where I worked with a person called Shewell, and he is co-author of my paper on holography you asked about. I have quite a bit to say about it. That turned out to be scientifically also extremely profitable.
I had other consults over the years. I consulted for IBM. I consulted for Franklin, I'm not sure of the whole name of it, it's associated with Franklin Institute, some title, Research Laboratory. I consulted there. There must have been other ones but I remember those best. I've been consulting for years with Schlumberger, which of course is a much more recent time. That's scientifically very profitable to, Schlumberger. That's an old prospecting company, old service company officially. I can write down for you some of these names later.
That's much more recent. It's the last six or seven years. But in the early days it was basically Kodak and Batelle. There were a few interesting things coming out of the consults with IBM. I have learned much more about the practical problems which are facing people who work with lasers and optical systems. But scientifically from my point of view it wasn't profitable. I don't know if it was from theirs, but from my point of view it wasn't, but that really doesn't reflect so much on IBM. These huge organizations, much of this depends on whether you find the right person in the company to work with. If the company will allow you to work with somebody, be reasonably free to allow you to work with somebody, it depends on whether you find the right person.
I obviously did with Marchant. I did not with IBM.
Bromberg: IBM would date from about when?
Wolf: The consults? In the sixties, early sixties.
Bromberg: Now, one of the questions I asked you on page 1, is, to try to understand how the laser community was receiving the work on partial coherence that you were doing, and I am guessing, you may want to correct me on this, that there was a gradual kind of interaction between the kind of work you were doing and other people were doing on theories of partial coherence, and the knowledge of the laser community. After all, people were coming into the community from all sorts of fields.
Bromberg: I said in my question that I had the feeling that there was some kind of tutorial process going on here. Now, that may or may not —
Wolf: — that is correct. And it was going both ways, of course. Of course. Well, you know, the very fact that I was invited to give some of the opening papers at some of the international conferences on lasers, like the one in Paris which you mention here, that was, I don't know the year about ‘63, ‘64.
Wolf: That was an indication of this because as I say, I have not specifically worked on lasers but I published a lot on coherence, and I was specifically asked to talk on that subject.
Bromberg: I see.
Wolf: It was in the opening section.
Bromberg: Yes, I know you were in the opening section. I didn't realize those were all invited papers.
Wolf: Yes. Oh yes, they were. There was a session with in it, I would have to look up the volume, but these were invited papers. And that gives you an indication; there was an interest in coherence, very much, because they realized there is more to coherence than just having something which looked very monochromatic. There is much more to it, and certainly — then of course there, that's a separate thing which I see you want to develop, but when Glauber published his first papers on coherence they had tremendous impact on the field, and then also the controversy between our group and him. Also for a long time I think it was very profitable for everybody because it clarified a lot of questions, in the long run it was very profitable for everybody.
Bromberg: Now, who invited you on that? Was that Bloembergen?
Wolf: I think it was Bloembergen. He was one of the organizers. He must have acted on behalf of an organizing committee.
Bromberg: Because he told me at one point that he was just accepting every paper that was offered, but this was not an offered paper.
Bromberg: Somebody must have —
Wolf: — no, that session was entirely a session of invited papers, and the one I remember definitely was on levels, another problem.
Wolf: But as I said, I can check. I may have correspondence on it. It's probably true that he was that he was accepting papers, there was a huge volume, but mine was definitely an invited paper, at the opening session. I probably wouldn't have actually had the nerve to go otherwise, you know, because I didn't consider myself a quantum electronics person. It was a quantum electronics conference. I wouldn't even have submitted anything for it.
Bromberg: There's a whole group of people who never would have — who never came to the first conferences, who suddenly show up —
Wolf: — that's right.
Bromberg: There's you and Mandel and Glauber —
Wolf: That's right.
Bromberg: Somebody was responsible —
Wolf: Oh yes. I was very flattered by that, of course. At least to me it indicated it was beginning to be a real interesting career, because before that, very few people were working in the field.
Bromberg: We might just go to that particular question at this point, even though it's a little bit far down —
Wolf: — could you occasionally give me the number of the question, because I have things which go to the question, so if I know the number I can look it up.
Bromberg: Yes. I wanted to ask about, on page 3 question 14, about how your paper was received there, and you have this idea of the origin of the coherence of lasers, because of the movement through periodic structures —
Wolf: Yes, lasers, yes. That was a very good discussion, after my paper. In Paris. Some of it is reprinted at the end of the volume. Somewhere in the paper there is transcription of the discussion, but actually it's not complete. There were many more questions. I particularly remember, because I remember it touched on the subject, Glauber accusing me that I have set optics 50 years back by using instead of…That for some reason has not be reprinted but it was part of the big argument we had at that point. That paper made some impact, judging from the discussions. There was a lot of discussion, not all necessarily positive. There was also some criticism of Toraldo di Francia, who saw the results, some of the results, on this publication, to be obviously not on calculations. I don't agree with him but the discussion wasn't all favorable by any means, but it showed that people were interested in the subject.
Bromberg: Because there was a lot in that paper. There was also the introduction of higher order calculations, and of course there were lots of different things people could have —
Wolf: Yes. You know, I wasn't the only one on higher order calculations, of course, around the same time three different people did it, Mandel and Glauber and I, because it was obvious the moment I started looking at the differences between laser and thermal light, it became obvious that higher order calculations had to be taken into account.
That clearly was the determining thing. Of course, the first higher order correlation experiment was done by Hanbury, Brown and Twiss. That as you know was a very key type of experiment. But the general interest in, you know, in higher order correlations, to classify them and so on, came with the laser. And to a large extent a lot of controversy about it.
Bromberg: Well, I'm going to go on back again then. You know, I'm back at 4. I think we've pretty well done 1 and 2 and 3, without any problems, and we want to talk a little bit more about these papers that you did. Now, Mandel had come for a sabbatical year here.
Bromberg: And you were working on these coherent light papers and —
Bromberg: Do you have any specific memories as to how you decided to work on these particular papers at this point?
Wolf: That's paper 42, and 43. Give me just a minute to refresh my memory — about this, I will probably —
There is this problem of some properties of coherent light, that's a paper which we published, that comes under this.
Now, you mentioned how, you asked the question how, G B Parrent's definition of coherence and so on. Let me tell you a bit of the background. You may not know that Parrent was my student.
Bromberg: No, I didn't.
Wolf: In fact, this is what I mentioned, at the Cambridge Research Center, Parrent went with me and he did his thesis in Manchester on partial coherence. As my student.
Bromberg: And that's why you had that Air Force contract?
Wolf: Let me tell you the story, because it — I told you that I had problems in getting a university job anywhere, it was very difficult in those days, and I had at that time a fellowship, called I.C.I. Fellowship, Imperial Chemical Industry Fellowship, and it was coming to an end but in principle it could be extended. I had one extension already, but I still didn't have a job, and there was a question of another extension and I didn't get it, and there was Parrent, only a few months after he came from the States specifically to work with me on partial coherence, and he realized that his thesis supervisor will be out of a job in a few months! He was always and he phoned somebody at the Air Force Commission Research Center, and they approached me, if I would be interested to open a contract for them, and they arranged a contract for me, through their European office in Brussels. Within a few months, I had support for another year or so, so I was able, you know, Parrent was able to write his thesis. Then during the year I had the offer to come to Rochester and he finished the thesis without me, and he did most of the work while I was there, but he needed to stay, there were strict residential qualifications and he had another six months or so to stay so he stayed another six months.
And during the time — see, this paper was published after I left. This is the paper which Parrent published in OPTICS in ‘59, and you see, it was submitted in July ‘59; it was already after I left.
Now, I'm sorry to say that there are mistakes in the paper, and I'm not the only one who noticed this. There are quite a number of papers in print where it has been stated. The results are basically right, but the question is, what and the mathematics is wrong, and just to show you it's not just me I should tell you that Parrent was one of my better students but everybody makes mistakes. Here is a book by Perina, COHERENCE OF LIGHT, and you will find a remark on page 51, where there is this remark, "elimination of zeroes (errors?) will be found in this reference, '' and then you look up the reference, this is a paper by, then there is another paper by
Baratat which appeared in the JOURNAL of the Optical Society, 1966, and you can read the abstract, and you see, examined by anyone reading it, you see that he pointed out also it's wrong.
I don't want to be too critical. These are pioneering investigations , some of these, and so on, and a subject advances only at its own natural rate, and the problem of complete coherence is much more difficult than people realize, and I'm still not — I can tell you a little bit more about that later.
At any rate, then we published some of these papers. Another one was this paper by Mandel, which became very popular actually. We got a prize from the Optical Society for it. They used to offer prizes for best paper published in the JOURNAL — for that year, for 1969, we got a prize.
But still, I wouldn't say that I would consider this the best paper we ever wrote together because afterwards we found it was not the whole story, far from it. It seems always you know, the subject must advance in its own way. There's so much more in this problem, of complete coherence.
So after that we published, years later, then I published a paper with Mandel on this problem in 1966. Then we published another paper with Mandel 1981, and I am giving a paper at the San Diego meeting in October, where I think I've settled the problem of complete coherence to all of us, I think now I completely understand it, but it brings in, technically it's quite a different sort of thing than people expected.
You see, there is such a thing as temporal coherence, spacial coherence; it turns out that there is also something known as spectral coherence, which is not so well known, spectral from the word spectrum, and let me explain to you why the problem is — and everyone overlooked it, and I must say, one of the things, where Glauber also was wrong in this, and other people published papers on this subject of complete coherence to all orders. Basically the thing is the following, that if you demand that light vibrations say at two points are completely coherent, completely correlated for all terms, the only conclusion that you can come to is that it is strictly monochromatic. It has to be a single frequency. And this is still — it's obvious that it will be fully coherent and then it's obvious that you will never have light like this in nature. You always have to have So it's academic, and in a sense, the way Parrent formulated it, it's a little bit academic, but the result came out which one would expect, that old argument, that if you demand it for all time delays, you have always complete correlations, it turns out to be monochromatic, doesn't help very much.
Now, Glauber looked also at high level coherence, and by the way, it's coherence to the second order, as he called it, to the most order the correlation function factorizes in a certain way, and Glauber quite formally introduced a factorization to all orders, to different coherence of orders But if the field is steady state, or in mathematical terms stationary, then already this condition in the second order implies monochromatic light, as I mentioned.
So the definition of coherence to the order on that basis, of demanding something, factorizing in the space time delay, produces definition of coherence which is useless in practice.
Now, I, a few years ago, started a theory which I called coherence theory in the space frequency domain, instead of time being the basic variable, it becomes frequency. It's much trickier because there are certain problems of defining frequency representation by stationary random processes, but anyway I got such a theory, and I have been applying it and extending it, and I believe now that I can show in what sense laser is really, a laser mode is coherent, and it's coherent to all orders, not in the space time domain but in the space frequency domain.
So I published it, the background is published, and this paper as I say, I'm presenting it at the meeting in San Diego. I'll have sent it; I'm showing it to you. I'm not saying this sort of to make publicity for this paper, but to show there is much more in full coherence than people realize, and there are a lot of inconsistencies in the earlier treatments, beginning with Parrent, and although Mandel and I published quite a number of papers, and I think I published papers with other people, it seems that this is not the way one should approach the coherence problems of laser modes. And it's much more intimately connected with this problem of — I don't know quite how to put it, but the question of the relationship between coherence and has been really completely misunderstood. For example, one can show, you can take an ordinary classical experiment, and put in front of the pin holes filters. Now, if you ask most physicists what will happen, as you are making the pin holes, as you are making the filters narrower and narrower, below the you know, they will say it becomes more and more monochromatic, so the sharpness of the fringes will go up. And it's not actually what happens. What happens is, you can see more and more fringes, but the maximum sharpness in the middle of the pattern has nothing to do with the And so I'm going to show these problems are a little bit trickier than people realize. To some extent I think it illustrates, will illustrate my answer to that question, and later on about the question of mathematical rigor and so on, why I am presenting some of these problems. One of them you asked about, holography, I will say something about later. The point is that what has been answered is largely the obvious question on a superficial level. Not in all areas — I mean obviously, the Townes Shallow paper is quite outstanding, but some questions connected , holography and so on , just skim the purpose, and the longer one works the more one finds there are things that one doesn't understand.
I think coherence theory is a perfect example of that. There's much more to be understood.
I'm sorry this became a little more technical.
Bromberg: No, no. What, I was going to move though to something that I want to point out about, there are really two things. I'll put both questions and then we'll sort of separate them out. One is that there is a kind of rapprochement that goes on between the concerns of laser scientists to understand the spectral distribution of their light, the —
Bromberg: — the line work and so on, and the work that people like you are doing, and one thing I'd like to do is to a little bit examine this rapprochement. Apparently the OSA meeting of spring of 1961, the subject of coherence was raised in some way or other; I don't know whether we can use that as a peg to the relevance of the coherence study definition to that point.
Wolf: Do you remember any people who were in this? It's so far back I don't remember.
Bromberg: To tell you the truth, this may be the wrong name, but is Neugebauer, does that sound —?
Wolf: — oh, I know Neugebauer, yes. It's not terribly relevant. You see, the traditional definition of coherence involves or timing an idealization but normally all the time responses involved in making measurements of light are very large compared to the basic optical periods. So, but with lasers it becomes a little different, so Neugebauer tried to modify the definition of coherence by in finite timing terms. We can only look at the net results, because now we are talking 23, 24 years later after this. He didn't make any — there was nothing wrong with the paper. I know exactly the one you are mentioning now. You realize it is an example of something I was trying to put in different ways, that there is much more to coherence than just, you know, earlier theories, that there are many more problems to look at, and one of them was the effect of very long coherence time of lasers on measurements. You see, if you would draw here there are two or three time scans involved in all these measurements. One is the problem of the bandwidth, that's one. One is the problem of averaging integer ?, right, and one is the basic period, so there are three time scales, involved, and now of three laser days in periods of days, they are in this order roughly, — this averaging interlude is always very very much larger than the bandwidth and certainly the bandwidth is much larger than the period. With lasers these two get inverted. You see, you can have a very very long coherence time and you average all those into it, right, and that's where one should be very careful, the limits one's putting on these averaging interludes, and that's one of the points which Neugebauer was stressing and it's obvious but his type of calculation didn't reach the answers to these questions. One knows now much better that time average, one should use — well, I won't say much better, but it's more appropriate perhaps to use statistical ensemble averages, and take into account properly the length of integration of the detector and this is something which of course Mandel has been doing very well.
Bromberg: OK, but I should say that in raising this question, what I really want to get at is, not the physics but the interest, if you can recall, or the way in which laser people were involving themselves.
Wolf: Yes. Sure. Well, that is obvious from some of the early Rochester conferences on coherence. I think if you looked at the second volume, I have a feeling Neugebauer in fact has a paper in one of them. I'm not sure, but he was in Rochester certainly, I know Neugebauer was — in the reserves?
Bromberg: Oh no, I didn't know that.
Wolf: In fact, a lot of it came while we discussed these problems quite a lot, and he worked in the reserves for many years. Then he went to Europe to represent some European country, I think, either Brussels, either in Denmark or Belgium.
So I know him quite well, by the way; when you originally mentioned it I couldn't remember it, but, he didn't settle the question, but it was exactly what you are saying. Great interest of course of the laser community in these questions.
Bromberg: In other words, one should understand that at a meeting like this where you have a laser session and somebody speaks on these things, people are really actively in there taking part, people like — I mean, who's in the laser community, Snitzer is a name that occurs to me, and the Townes group, and the Hughes people I don't know, Hellworth and —
Wolf: Yes. That is certainly right. But you must remember that many of these people are much more interested in engineering applications of lasers, and providing lasers, and I'm not saying that as criticism, but engineers have very different interests from say theoretical physicists. They know that the work of theoretical physics has quite a lot of impact on what they can do. But I wouldn't say that that large number of people who worked with lasers were really interested in some of the rather basic questions on this.
I'll give you an example, actually it comes later but
I'll mention it now — special of laser modes. You know, the basic classic theory of laser modes is due to Fox and Lee, and it's a very important theory, but from the theoretical point of view very primitive. I'm sure they would admit it themselves. And it leaves a large number of questions unanswered and one of them is to prove mathematically existence of a single mode. It's not a simple problem, because there is that which is associated with the propagation which turns out to be non-emission, and there are no general theorems on Eigen functions and Eigen on non-emission curves. And it took mathematicians years before they proved that a Fox Lee equation has a single solution, single mode. There are three papers now. I could find them for you if you were interested. And mathematicians have not been able to prove at all that the set of modes even for simple geometry is complete, in the sense that completeness means basically in the context of laser physics that you start with any initial distribution and eventually after a lot of bouncing inside, it really produces simple position of modes. This has never been proved.
Now, you may think this is academic, but then it comes to unstable resonators. People are arguing to this day whether unstable resonators have modes and in what sense the modes should be defined.
You know, as long as a theory works, you know, works in the sense that you can test things in the laboratory and it seems to get — most people in the field, this is not just engineers but many physicists too, will not bother about the foundations. They start worrying about it when something goes wrong or when there is something they can't understand.
Bromberg: That's nice. You see, what I'm really trying to do here is to use your answers and also the answers in many other of these interviews to begin to get some sense of the articulation of the laser community. That's why this question 5 goes right into a question about the professional world or the invisible college, that you were in contact with, because one of the things I think people are going to be interested in, in these interviews, is what were the diversities within the laser community ? What were the sub-communities?
Bromberg: Which, for example, the kinds of specialists which y ou were in contact with and the kinds of people who were interested at various points as we go through the sixties , in the work, and the kinds of crises that did or didn't emerge, that brought coherence to the forefront. So this particular question sort of was in the early — so I'm getting a feeling that maybe DeMaria and Sigmund and McBertry, people like that, would not — of course, Sigmund and McBertry were going to be worried about coherence, weren't they?
Wolf: Yes, amongst the few I have mentioned, because Siegman has been very much worried about modes of unstable resonators. In fact he was very much on my mind when I was answering this question. I have actually notes of lectures which he gave, one of these short courses at an Optical Society meeting like the one I'm giving in September. I have the notes from a course he gave a few years ago, especially on unstable resonators, and he specifically questions, I have still in my mind the transparency, a big question mark, existence of modes in unstable resonators.
Now, that's sort of the next step. Once one understands the modes of an ordinary resonator or a stable one. And as I say, many engineers and many physicists would not be bothered, how to prove them. They are there, right? You make, the laser ray can obviously prove their existence. It's a different thing, and some people may say it's academic, but I'll give you an example later from holography to show you that, ask me some of these questions you want to know about the holograph later. It actually produces new physics, and very interesting physics.
Bromberg: I see, so Siegman is an engineer, as you say —
Wolf: — but with a very strong feeling for basic questions, much more so than many other people in the field.
Wolf: By the way, I consider his book on lasers one of the best books. It's fairly elementary, but he asks the right questions very precisely. You know his book?
Bromberg: No, in fact, I'm going to read it.
Wolf: Very nice book.
Bromberg: OK. What else, so then, we have now beginning to get a feeling for what your scientific universe was like at this point, and the people at Xerox and Eastman Kodak , and Mandel of course and the people right here . Are there other contacts that really one ought to understand in this? Were you very much in contact with European scientists through this, —?
Wolf: Well, to some extent, not greatly, but I don't think you should consider me in some sense to be typical person in the collective laser community. I repeat again that, important as the laser is and I realize, you know, lasers are not my primary interest. I've been largely influenced with the questions, asking questions, on lasers, also by my contacts with Mandel and my colleagues here. But I've always been more interested in some basic questions in optics, the basic questions like diffractions, image formations and so on, so I don't think I would be typical, if you want to draw some consequence of this. I was in contact with many scientists, many optics people, but not specifically on the subject of lasers. I was in contact with often with this man who is a professor at Polytechnic who wrote a very very fine paper on this question , basic paper on this question of relation between radiometric and coherence. I've been in contact with a number of European scientists, but of course my main contact in Europe was always Mandel and of course he came here, because we shared similar interests, but I wouldn't say there are many many people I was in contact with on lasers.
Bromberg: Now, there's just one other thing in this whole group of questions I want to say. Some people I have found, found that their scientific monitors in the Air Force, for example in your case, were very important scientific contacts. I'm getting the impression that was not true for you, but I just want to pin that down.
Wolf: Important in what sense? They got me in contact with other people, you mean, or?
Bromberg: Well, in fact I was thinking of intellectual importance, but if that, if they had social importance, that would be worth speaking about.
Wolf: I must tell you, it sounds — well, it doesn't sound terribly good, but the main item of usefulness also, but the main point of these contacts were that these people were steady, they were willing to continue to see to it that I was supported and that I could continue to do the research and have students and so on. That really basically was the main thing.
Wolf: Marginally though I should say they occasionally got me in contact with people who would send me reprints or preprints or something, which they knew I would be interested in, that they were also supporting usually. That was very helpful. That's about it.
Bromberg: That's also interesting, getting the role of these kinds of military research agencies is something we want to know about. OK, well, question 6 is a question that happens to interest me a great deal.
Wolf: The signal. You will be very amused by this — well, I don't know, amused, but I show you this. By the way, I have (Dennis) Gabor use of his picture in my study.
Wolf: In fact, it was through Gabor that I got to know. That's how I got to know
I wrote in 1955 what I consider the basic paper, amongst the ones I wrote, from my point of view the basic paper on coherence, and I would like you to read a footnote on page on page 248. Could I show you the footnote?
Bromberg: The one that says "Since this was written, I find the same complex representation of real fields, where it was introduced previously by Gabor in his interesting investigations in communication theory –“
Bromberg: That is interesting.
Wolf: Yes, I rediscovered the erratic signal. I didn't know that the real imaging process involved the transfer as Gabor showed in his paper which I refer to in 1946. I'd known Gabor for years but we never had discussed this. And I was forced to introduce the erratic signal essentially to have a really consistent elegant theory of fluctuating fields. So when you talk about the influence of the language of communication theory and so on, sure, I knew of many of the papers in communication theory, but I was actually forced to provide this, to develop this representation, and I was amazed when I discovered that Gabor had done it years before me, and in fact had it by putting, deriving explicit relation between the real and imaginary parts. It's rather interesting also in what happened afterwards, because you know that in its theory of course the starting point is the decomposition of the field into positive and negative frequency parts, and that is strictly analogous to introducing a signal.
Now, Gabor claimed in his papers that one is forced to it in quantum mechanics. I say it's a matter of mathematical elegance. I would leave this an open question. I don't think it's an accident that Gabor found and I found that one really has to use this type of language to make a consistent theory. I think it's even now too early to judge that. It looks very much like a matter of convenience in the classical theory, but I don't think it really is, and I think that Gabor would be the first one to claim that it's more basic than that.
But again, I'm sorry if I disappoint you with this question, but it was not the influence really.
Bromberg: No, I think that's very interesting. It was the influence then of the subject matter which drove people to do comparable relations.
Wolf: This was in the very early days of my theory. This was essentially my second or third paper on basic a proper theory; I didn't even know it was called that. I called it here complex half range functions, half range, only half of the frequency component, you see. And then I discovered that it was known as the signal by Gabor, later.
Bromberg: That's an interesting answer. Now, I also, 7 no longer relates to 6 but I want to understand anything that you saw about the way the laser influenced the Optical Society.
Wolf: Oh, very much so. I am not sure which year you particularly are talking about.
Bromberg: Well, I'd like to start at the beginning. Now, you came in in ‘59, so your observations.
Wolf: No, but I was actually very much involved in the Optical Society a little bit later. In the late sixties and early seventies, I was on the board of directors of the Society and I was president of the Society, so I have seen quite a lot about what happened in the Society, and the impact of the laser was tremendous on the Society.
Two examples which I put down here because you asked this question, one was the formation of the Technical Council, which didn't exist before. That was a part of the administrative or scientific part of it which was responsible for setting up technical conferences on specialized subjects. Before that, the only meetings of the society were really two annual meetings. But the laser had a tremendous impact on the Society. People started joining the Society, you know, who certainly were not interested in ordinary traditional optics. And very soon people started becoming leaders in organizing the technical part of it. Of course A.J. DeMaria, people like DeMaria. There were many others but his name comes very much to my mind. And I would say that the Technical Council was a very very important part of the Society, and its organizing meetings the Society, three or four technical meetings a year on specialized subjects. Most of it, you can see in it the influence of the laser on it, practical, you can see that.
So those were the main influences, I would say, on the Society. I mean, the manifestation of the influence from the laser, the manifestation of it was the formation of these, the formation of the Technical Council, and running of these topical meetings.
It wasn't without its problems, you know, because I remember quite well the number of people who worked in traditional optics who were afraid that the laser people sort of would just take over the Society, the whole Society. They thought it belongs more to?
And there is a lot which went on behind the scenes, you see. The editor, I don't really want to go into names, you know, to be critical and so on, but one of the editors in those days was not too happy to accept papers on that subject for the JOURNAL OF THE OPTICAL SOCIETY OF AMERICA. He felt it belongs to engineering journals. As a result of it, there were practically no laser papers in JOSA for many many years, and you know, in more recent times the Society tried to change this, and they went really out of their way to do it, and one of them was starting this new journal OPTICS LETTERS. OPTICS LETTERS was precisely to attract people from the laser community. I already was on the board of directors then so I know the history of it. It was specifically to attract people in quantum electronics and the laser community in general to start publishing in optics journals. They thought if it goes through OPTICS LETTERS, then there's a better chance you know of getting them to publish in JOSA, and it didn't quite succeed, and as a result of it, the JOURNAL of the Optics Society was divided last year in two parts, you know, Part A and Part B.
Bromberg: No, I didn't know that.
Wolf: Part B specifically devoted, they call it optical physics, but the idea is to attract as many laser papers in it as they can. So there is still, you know, this system continuing this process of this influencing — well, originally I would say it was sort of infiltration of a lot of, you know, engineering type of physics, leading to the Technical Council, but it didn't catch on with regard to publications, and OPTICS LETTERS and now the JOSA, they call it B, is an attempt by the Society to get this right.
Bromberg: But you have APPLIED OPTICS starting in ‘62 which is really —
Wolf: — sure, there are many applications, yes, but basic papers, I mean, more basic papers on lasers are not published there. It's specifically applied optics.
Bromberg: I see.
Wolf: And many certainly physics people, most of them don't read APPLIED OPTICS. I have nothing against it. To give you an example, you know, these journals most of the members used to get free, but now there are so many Journals, the Society can't go on like this, so people are subscribing to Journals. They get them at reduced rates. And I don't think anybody in the physics department subscribes to APPLIED OPTICS. Many of them subscribe either to the JOURNAL of the Optics Society, B, or OPTICS LETTERS. But the basic , you see — APPLIED OPTICS is really what the name says, it's applied optics, nothing wrong with it, but it's not the sort of thing I sort of felt we were trying to explore here, some of these basic questions , collective modes, collective coherence, bandwidths and so on, by and large that doesn't go into APPLIED OPTICS.
Bromberg: On the other hand, it does indicate that somebody in OSA, somebody in the Optical Society was interested in lasers, that they would even put this magazine out, which very swiftly put out a big supplement on lasers.
Wolf: Well, you know, it would be interesting actually to look through the first few volumes of APPLIED OPTICS, to see to what an extent lasers were really influenced it.
Bromberg: Well, I think question 8 on the motivation for higher correlations has really been answered. One thing that I notice in question 8, and of course these questions will become part of the record, is that I ask you here about the principal steps in novelties and obstacles and breakthroughs, and in general, I do think that we ought to pause on some of these works to pick that up, where I've sort of ignored this as little bit, but we ought to be sure as we understand your research on coherence, and I'm kind of leaving out your research on diffraction, to see where the principal breakthroughs were, the principal false starts, or what was actually going on conceptually. We ought to be sensitive to that.
Wolf: All right. Now, as you mention that I became aware of the importance of higher effects quite early on, in the first experiment on elementaries, and that again has a strange coincidence because the work was done at Manchester by Humbery Brown?
I was at Manchester from 1954 to 1959, and Humbery Brown was at the Jodrell Bank Registration, which is part of Manchester but just outside, and I knew him. And through him I got to know Twiss. And I don't know into what details you want to go here, but I learned about these experiments for Humbery Brown even before they were published, and once he started to talk about them, people got extremely negative, particularly people who had were exposed to fundamental quantum electrodynamics. (?) And I can remember specifically, there was (Leon) Rosenfeld, a very famous physicist, (crosstalk) —
Bromberg: — I worked with him —
— yes, you worked with him, and you know, shortly before Hanbery Brown, sometime before Brown started publishing his papers, Janossy in Hungary published a paper pointing out that there are no such correlations of the type which of course were later found. And in spite of that, Hanbery Brown went on in this, and he was larger an experimentalist, with very strong intuition, and he was a friend of Twiss, a very good friend of Twiss, and between them they did very good work, and I remember still his telling me direction of radio waves, why shouldn't it work with light?
He said, "I don't understand all these in these books but it seems to be the same thing, why shouldn't that —?''
Also in these experiments, there was Podolanksi, who died six months ago, Podolanski around that time was still alive in the early days of these discussions (or maybe it is) Podolansky died shortly around that time but he was still alive in the early days of these discussions — didn't believe it could work.
Well, I mention this partly because I happen to have published a very short note around that time, in PHILOSOPHICAL MAGAZINE, I would like to show you that one because it’s also...later, I don't think now, but —
A very little note published in 1957, in PHILOSOPHICAL MAGAZINE, the purely classical calculations, but I showed that there should be, on the basis of classical theory, and invoking what is now process, this is basically a forerunner of correlation effects. It was the first time I was aware of how the correlation effect was and it just happened that for Fermi light, on which they were doing experiments in those days, one could explain it, one could reduce the first order correlation to second order correlation. And I mention it here because it's also the — related to subjects we will discuss more on this, other criticisms.
At least at the time I was aware that there are other higher correlation effects which correlate with some experiments.
Now, Rosenfeld didn't believe it. When the experiments were successful, we had many discussions and I believe that I was the one who explained to Rosenfeld, but I still don't know what else actually but was explained to me by also.
What it is that the quantum mechanics are so subtle that even people like Rosenfeld, who did some famous work as you know, couldn't understand it? Not as a real problem. The first person who really understood of course was in this country, Purcell — you know, that famous — he wrote that famous piece in NATURE with But now to pretend that all this was obvious or trivial and so on, from the point of view of how the — and the history of the problem shows it.
So, around about 1957, I became aware of higher correlation effects. And then of course main impact came of course to Glauber, and in spite of our criticism, of course much of what he does is great great work, and it got me interested in higher correlation effects. He wasn't right in everything. As I mentioned his definition of higher order coherence I think is, not useless, is probably the right thing to say, because in the controversy in Paris, Senitzky and Glauber had this heated argument. He said, "You know, the Glauber definition, how can Wolf and Mandel's definition be wrong, and the definition a definition? It either applies or it doesn't. ''
Now, it's the same sort of thing. I would say Glauber's definition of higher coherence, complete coherence, higher complete coherence is useless. But that is, you know a separate thing. He did very great work, and it was really largely, in my attempt to understand many of the things he was doing, that I got interested in higher coherence which is of course a very important subject.
Bromberg: I don't see why we shouldn't talk rather directly about the controversy and the work coming out of it at this point.
Wolf: Well, if you want to.
Bromberg: Then we'll come back to some of these questions.
Wolf: All right. All right. You know, we published, Mandel's paper in the PHYSICAL REVIEW which you asked about. That was, let's see, — you asked about this paper —
Bromberg: That's now 43, number 43.
Wolf: Do you know which question it is? Unfortunately, I think I put it into —
Bromberg: I see (crosstalk) I'll get the question out. It's Question 15.
Wolf: Question 15. Yes, Now, our whole controversy with Glauber, and I say ours, it started with Mandel and me on one side and Glauber on the other, and then of course Sudarshan got involved, I think it was the publication of our paper in 1961 in PHYSICAL REVIEW. This paper was largely concerned with correlatins in photo detectors when the photo detectors are illuminated by light which can be diffracted by the Gaussian random process. This is explicitly stated. There's nothing wrong with this paper, but there is a sentence there which is deceiving, and I agreed that it was not very clever of us to do it, — we made some remark about lasers there which actually, it seemed that most lasers were characterized by Guassian processes, but if there are several modes, even five or six, it's already a Gaussian random process, you know, the stuff is —
Glauber jumped at this remark about the relation to laser, and of course if you interpret it in the sense of, that with a single mode laser you would get this effect, he's absolutely right, it doesn't apply. We were not careful enough, but if you read the paper, still, it still is — there is a Gaussian random process, and in hindsight of course everything is very easy to understand. This was ‘61. No other type of light was known except that which was governed by Gaussian random process statistics, but basically it’s misleading.
Well, Glauber jumped at it, and that started the whole controversy. You have to understand he's a very aggressive person.
As anybody can tell you who knows him. He's a great scientist but aggressive, and he started attacking us very strongly in print, in a paper he published in PHYSICAL REVIEW I think or in PHYSICAL REVIEW LETTERS, and then of course at the Paris Conference.
Bromberg: By the way, the conference in Paris was February,’63.
Bromberg: Then Glauber and you must have already had this exchange, because you have a little paper, I think it comes out in February "63, answering Glauber.
In '63 was it? I'm not sure. But we became aware of Glauber's paper when it came out, and I think it appeared just before the Paris Conference. Look, you know, I really don't remember without the vita, it is just possible that they sent it to us for refereeing, that we actually saw it before it was published, and we certainly didn't block it, but I am not sure of the details anymore. But we were aware of that paper very very early after it was written. I don't know how, I think probably refereeing. So we were prepared that we might have a controversy in Paris already. But —
Bromberg: Your papers for Paris were probably already written at this point.
Wolf: Yes, that's right. That's right, yes, but, so we were aware what may come, but what I'm trying to say is that Glauber jumped on one sentence in this paper, and he's right if he interprets it, because logically for one mode the Gaussian random process doesn't apply, but the paper is concerned that there should be more modes later.
But sometimes as you know controversy is very healthy. People got very much interested in this controversy because we were reasonably well known in the field, and here's somebody else pointing out that, you know, much of what we have written seems to be turned around, and now Glauber was disguising all this in the language of quantum electrodynamics. The basic concern has nothing to do with quantization, it's whether you have Gaussian statistics behind it or not. It's true that if you want to go to the limit, you know, and do the finest type analysis of all these experiments, you have to bring in finer and finer theory, of course, but you start in classical optics, you can start with scale events and for a more complicated problem you take electromagnetic waves, and if it's more complicated you apply ordinary quantum theory or semi classical theory, if it's still more complicated you go to quantum electrodynamics.
I mean, you have to match your problem with how it's treated. Now, Glauber gave the immediate impression that what is needed for this is quantum electrodynamics. That is not true. In fact, to this day there are only three or four experiments, mainly Mandel's and so forth with statistics that use quantum electrodynamics. Most of it can be done with classical and semi classical. But it stimulated a tremendous amount of interest in the field, which was good in the long run, of course. Of course, many of the more subtle questions were really clarified by this.
So, don't get me wrong, I think extremely highly of Glauber's work, but I think he got to be very well known by what I would consider — well, to put it mildly, not exactly gentlemanly types of procedures.
He would have got the same publicity eventually because he was very good without all these attacks.
Bromberg: All right, but let's go and see how in specific ways it sent you or your collaborators in specific directions. If we can somehow go back and reconstruct the intellectual journey.
Wolf: Sure. It did a lot for us. A great deal As soon as we can from the Paris Conference we talked to Sudarshan about it, who was our colleague here, Sudarshan was in Rochester, and Sudarshan by the way is a very broad minded physicist who looks at a problem from every possible angle. He's also a friend of ours. We could talk to him very freely and easily about the whole thing, and he studied it for a few days and produced his representation of his views on this controversy also with Glauber which was essentially a quasi-classical type of representation of quantized fields, so then there were three of us, arguing against Glauber, in a way.
Bromberg: This arguing was being carried out by the papers or telephone talks?
Wolf: No, no, no, arguments with Gluber, you mean? No, that was all in published papers, very little contact between us personally. In fact Sudarshan was offended by some of Glauber's papers and wouldn't have anything to do with him personally. We were on speaking terms, but I always thought that he was rather rough. Glauber has a bit of a reputation for that, but still, let's stay to the scientific part where Glauber was very good, stimulated interest, and he certainly influenced all three of us, Sudarshan, Mandel and myself, to a large extent. In my case it was greater interest in higher correlation effects, of course.
Bromberg: I see.
Wolf: Yes, that was the main thing, and I'm still writing on them, — as I said, I have new results which I do believe will...
Sudarshan was developing this, what he called, well, people call it re-discussion or presentation, some of them, I would call it peer presentation. There were peer presentations to Glauber. It's the same thing that Sudarshan arrived at independently. There is the question still even now of priorities but they arrived at it independently around the same time...
Mandel, of course, I think, I'm not sure, you should ask him, but I think many of the experiments he's doing to this day about finding limits of classical, semi classical methods, he was influenced also by this controversy. Because it turns out, after all this, there are very very few experiments for which the full coherence theory quantum. is needed. But he is much more competent to talk about that.
Sudarshan would probably say the same as I'm telling you now, that practically every problem in optics, which turns up in optics, can be treated by classical or semi classical methods.
Mandel is a little more cautious in this and he can produce of course examples which can be, and he's right, his recent experiments — but there are about two experiments, two kinds of experiments.
Bromberg: Now, what decided the three of you to collaborate on this paper that you did together.
Wolf: The one on the photoelectric light fluctuations? Yes, I'm pretty sure it's the paper; let me see... that was in question 15, 56, yes,
Wolf: Well, that paper — refresh my memory about it... I don't want to say anything... the dates, 15.... I must have taken the paper out... I think it was before...
Now, you see, Glauber "s starting point really was the description of how a photoelectric effect really revealed something about the statistical properties of light. It's exactly the same problem as Mandel started working on, in the paper by Purcell about 1956. Mandel was in on the same thing about quantum electrodynamics.
Now, this turned out to be very important also, to find out whether Glauber was right about claiming that the semi classical theories don't work. It was also extremely important in connection with Mandel's work, because Mandel's who theory was semi classical.
Wolf: And Glauber tried to do this by quantum electrodynamics. Well, this paper I'm talking about now, that's our joint paper with Mandel and Sudarshan which was published ‘64, and I was directly resolved to see how far one could go with a semi classical theory.
Bromberg: So it's really a kind of a, well, crisis is probably too strong a word, but you really had to get to the bottom of his criticism.
Wolf: That's right; because you see by this time it was not just Glauber. There were some other people who supported him, of course. Particularly there was a fairly sharp one by (Paul) Kelley,
from Lincoln Lab, and I see also (Hermann) Haus from MIT, who were doing the same sort of thing as Glauber by quantum electrodynamics, but they had never looked at the opposite, how far one can really go with the semi classical theory.
Bromberg: I see.
Wolf: So there were quite a few other people involved, not just the main papers, largely, one or two papers which we published. Well, you know, this is not the sort of thing on which I am an expert, but it was something which I was very much interested in, this problem, and there's part of it here. My main part was the subject and the information to it. (?) I was very much influenced by description of Einstein's fluctuation formula of 1909, which is not terribly well known by physicists actually. It's a very interesting —
A book by of course, by Born, this chapter here, let me see, it's a masterpiece, "Mass, Energy and Radiation.'' And he really discusses Einstein's work on the wave particle duality published in 1909. Einstein did a lot on wave particle duality but this paper is not very well known because it appeared, I think it was the result of a conference, I'm not sure.
Bromberg: It is very well known by historians, by the way. One of the key Einstein papers.
Wolf: It's a very beautiful paper, and essentially what Einstein shows there is that if you look at energy fluctuation in a small region of an arrangement, with a frequency range, the energy fluctuations, if you work them out, the variance of them, based on Planck's formula, gives you two terms which Einstein interpreted in a most ingenious way. One of them is the result you would get if a black body radiation consisted entirely in terms of classical particles, and the other one is classical waves but side by side, you can also say that it illustrates that the classical particles there which are a very different type of fluctuation formula, classical waves then.
Now, of course this was also the starting point of Purcell's analysis of the Now, I was very much interested to find out whether the formula could be generalized. It was derived by Einstein from black body radiation only, and this whole controversy around the Hanbery-Brown effect, is it particles, is it waves, is it both and so on, somehow one has to understand it in that language. So, when we were writing this paper, much of it is the consummation of Sudarshan and Mandel, I was keeping my eye on this fluctuation problem, and this part, which is towards the end in section 3, where we showed quite generally that the fluctuation from any type of beam, the variance of the photoelectrons, is a particle structure, a wave structure, but there's a constant here, and this can vanish in some cases, it can vanish in the laser case —
Bromberg: — I see —
Wolf: — but in general the two terms are always there. For me this was the most satisfactory part which came out of it. You can do it now by quantum electrodynamics also. So then you ask about the background of the problem. We started trying to see whether one really has to quantize the field completely to be able to treat it by quantum electrodynamics as Glauber claimed , or it could be by semi classical arguments. We started by trying to construct a wave function of photons in configuration space which gives a lot of cover. Mandel will tell you more about it tomorrow because I found a whole our whole correspondence, I mean, Mandel's description never published but related to this .I gave it to Mandel because I think he's the best person to discuss this part of it.
And as I say, my main interest was to see whether we could get the fluctuation formula out of it, which finally came out.
So the background of this paper, to put it in a few words, was to understand whether Glauber was really right in applying, in insisting on applying quantum electrodynamics to this. The answer seemed to be, no, you don't need it. And this was really the answer.
But if you people are interested in these basic questions, by now, I don't think anybody would actually use the fully quantized theory to treat these types of problems, other than Mandel has been doing since, he used similar methods, he provided a full justification for the things Purcell started, and Mandel was doing it in the fifties in England. He provided a very sound basis for it.
Bromberg: When you say you don't think many people are that interested, does that mean that you only had a small group of people who were reading this and commenting on it?
Wolf: Well, look, the only people who would really be interested in, do you need to quantize the whole system fully or do you only just, can you use, just quantize the field , and use the classical, — I don't think many people are interested in these questions. They are interested to know, what is — you send a beam of light on it and you do some calculations and give us a formula to work it out, and the formula looks the same whether it's classical or quantum electrodynamic or semi classical. Well, semi classical, classical doesn't quite really —
Bromberg: — what I'm really asking you is, do you happen to remember who got involved in this particular —
Wolf: — type of controversy?
Bromberg: Well, this paper, who wrote you and said "What an important paper, I must have a reprint '' or whatever.
Wolf: It was quoted quite a lot afterwards. But people like say Haus and Kelley and Glauber, I don't think were the least interested in it. They would prefer to do everything by quantum electrodynamics — if they can do it, fine, but there is a limited number of problems you can solve by these very powerful methods, you know, legitimate problems. As I said before, you have to fit the problem to the theory.
Bromberg: This is a kind of a Cambridge School that you had.
Look, many papers are, educate the writer. From that he can go on and so on. We have all learned very important things from this paper ourselves. And the satisfactory thing was that Mandel's treatment of '56 stood exactly as it was. I found it very satisfactory because the paper, particularly the one by Kelley, and he was not the only one, I think it was a joint paper by Kelley and Kleiner.
Yes, that's right.
Wolf: He produced Mandel's formula by pages and pages and pages of long calculations, and it looked at the end exactly the same as Mandel's. And this provided the justification for what I say. But whether you can convince people who really are used to that type of language that the things can be done differently, I don't know, but I personally found this a very satisfying paper.
Bromberg: And am I right to understand that Mandel was in London at this time, when you were doing this by correspondence?
Wolf: Let me tell you that the finishing touches were put on it at Mandel's home in London. Sudarshan came and we were up at 3 or 4 in the morning, and we still talk about — Mrs. Mandel on this occasion was serving us drinks and coffee all night, while we are finishing it. So that means that Mandel must have been still in London. Sudarshan was in the States of course and I was there, but — I was here also — but it was finished in London. But it started differently. It started by trying to construct essentially photon wave function considerations basically, and I would like you to ask Mandel about it, because we found a pile of papers from the early days, from this.
Bromberg: Good. OK.
Wolf: You know, let me mention just one more thing, one of the key formulae in this whole thing is formula 21 which Mandel produced by quasi-classical arguments, and it's sometimes known as modified or something distributions, average distributions.
The key to the whole photoelectron statistics and no subsequent work has changed this formula. Mandel wrote it in ‘56 and it is still true today. I understand, you go back to the origin of it (?) but it's still true. It's probably I think the most important thing he did.
Bromberg: I want to ask you one more question. There's something that I'm getting wrong. I think that, I keep asking, what were the obstacles or the breakthroughs or the ideas you had to give up in order to get there. I still have misunderstanding what this kind of theoretical work is about. What happens when you want to explore a question like this? Is it a matter of being hung up for weeks on an idea and then suddenly getting a gestalt switch? Or is it —
Wolf: — that can happen. But more often it's that you try to understand a certain thing. For example, this question of (crosstalk) laser modes —
Bromberg: — you do the populations more carefully —
Wolf: — and you try it and you do it more carefully and it may not come out. You try another way and you try another until you come to it. Sometimes you get a sudden inspiration and you see how it's done. But it may take decades. This business of generational spacial coherence through periodic structures which started in Paris, I really believe that I only answered a few months ago. We have a paper about to be published and I'd like to talk to you about it. Because, you see, the problem of coherence of laser modes is one of these questions which have not really been treated until now. Properly. People just know that laser modes are very coherent, but if you ask them exactly in which senses, they will start
"Well, it's spatially coherent, of course, the laser somehow — it has coherence time,'' but what has that to do with complete coherence? That's a finite bandwidth. You also have always finite coherence time.” You have to make interference experiment over time, and bandwidth, so it's certainly temporally completely coherent, right, so at most it's spatially completely coherent, and there is no proper theory which separates the one from the other, because spacial and temporal are related. It turns out that this coherence the earlier spectral coherence, and laser modes are coherent frequency by frequency. In a different sense than was understood before. And it illustrates this point, that it takes years and years sometimes, you ask the question, you answer it, but then you find you didn't answer the whole question or some aspect of it. You go on and on to another aspect.
Now, tried to answer it at the sort of lowest level which as appropriate to that time, but its complete coherence in the space-time domain, and you've got monochromatic light out of it essentially. It's nothing surprising and it's not terribly deep. It's a starting point. And then you find you have to put something more realistic then, so we did that, Mandel defined the coherence via the maximum value of the particular coherence, and that had illustrated something but it was still not the full story.
I believe now it's settled, but it's very recent. And it illustrates I think this gradual evolution of an idea, you know, and something, you keep bouncing it, something to do with it — that’s of course what already Fox and Lethery realized, you know, the basic interpretation. It started as bouncing and went on to as bit of coherence in, and then one wanted to do more and more. One found that one had to sharpen the concept of coherence to be able to understand this.
I would like to add, you know, that understanding in physics is very subject to this. You can talk to a number of people, and ask them the same questions, and they will say it was fully understood 50 years ago — and in some sense, they're right. And I will say it's far from the whole story, there's something more to learn about it, and I can show much more recent papers, problems which the old theories can't answer.
You may come to it later when you ask this question about — which you ask in the written questions — about my scientific style, or about holography, which I'm very keen to come to.
I hope you understand what I'm saying. There may not be a unique answer to this, what understanding really means, you know. Even in basic physics, you start asking, you finally come to what you think are the basic questions — in what sense do we understand really the basic?
Wolf: So there are levels of sophistication, and my level of sophistication, the coherence of lasers has not been solved until the last few months. But if you talk to somebody else, you'll get a different answer. We should take a break.
Bromberg: We're returning now after a short pause.
And we want to take up some questions that you picked out as of particular interest.
Wolf: Well, I should say of particular interest to me in several ways. It's very subjective but I think one can learn a few general things from it. You want me to pick them out, no? Well, the main one I'd like to come to is your question 19 about holography. Let me look at it carefully...
You say that my own work on holography comes after the first paper by Leith and Upatneiks was published, and how did I come to take up this problem? And it also ties up then with the question which you ask about my papers being in some sense more mathematical than many others seem to be. Now, I'm really interested in this question for several reasons, and also — you're absolutely right that my paper on holography was published in 1970 which is many years, at least eight years after, six or eight years after these...
Now, the first thing I want to stress, is, that paper did arise actually from my consultancy with Batelle. They essentially gave again, as some other companies, gave me a free hand to collaborate with somebody there on a subject which seemed relevant to their work and surely was very interesting, holography, so we started working on it.
I told you that I got a letter from different people who think they solved the problem. There are more questions left than answered, and holography in my opinion was one of them, and I wrote, I should say, not just this one paper, but I wrote three papers on holography. This one you just mentioned, but there are two others by me alone, just around the same time. I had a paper on
holography in JOSA in 1970, and another one in OPTICS COMMUNICATIONS in 1969, so that was the time, around ‘69, ‘70, when I really got into holography.
Now, the first question which I wanted to understand, and which had not been answered until that time, is the following one: a hologram, as you know, if you properly illuminate, you reconstruct the object, and that means that the photographic plate on which you have the hologram must have recorded information both about the amplitude and phase. I wanted to know more precisely how the information is stored into the hologram, and how from it, mathematically, one could reconstruct the three dimensional object, because this is a three dimensional object.
In that sense the holograph is completely unique, it stores three dimensional information, in the true sense, not just perspectives, but really with regard to the physical content of the body that is stored in the hologram.
So that was the question I tried to understand, and these three papers are all connected with it.
Now, I am a firm believer that if one solves a problem correctly and properly and goes into enough depth, which one gets answers to questions one has not really posed, if one goes far enough, and this is a clear example of it. This little paper I published in 1969 which is entitled "Three Dimensional Structural Information of Transparent objects from Holographic Data'' , it made no impact on holography, but actually, I give some references later, it was the starting point of what is down as diffraction tomography. You know about the CAT scan?
Wolf: Now, the CAT scan, which is done with ordinary X-rays, but if you now start using say ultrasound waves, which are sound waves, much longer than X-rays, you cannot use the same physical principle and the same mathematical principle for the reconstruction. The X-rays, you can essentially use geometric optics, X-ray goes through an integrating and....
Bromberg: We were talking about records we would use or not use. So then I was asking first of all about the circle of problems out of which your first coherence studies arose.
Wolf: Well, I worked with Maxwell in Edinburgh from 1951 to ‘53, we continued our collaboration later on, but during the time when we worked together on the book in Edinburgh, when I came to write the sections on interference of light, I was very dissatisfied with the usual treatment. It seemed to me that the basic problems had not been properly formulated at all, and so I started looking around, what else is in the literature, and I found that apart from this great idealization which most textbooks use when they talk about interference, using essentially a monochromatic model, there were few people who looked at the problem on a deeper level, bringing in statistical concepts. And the most important one undoubtedly was the work of in 1934. There were one or two papers before that, particularly by Franz Sitter, and even earlier than that by but they had been largely forgotten.
There was more current interest in the subject by H.H. Hopkins of Columbia University, who made fairly substantial contributions to coherence theory but mainly in applications to instrumental optics, that were not I would call the statistical foundations of it.
So essentially I started working on it myself then, having started with paper and developing the theory, so that was really the origin of my interest in the subject. To summarize it, I wasn't satisfied when I was writing the corresponding chapter of the book that the usual treatment is right or adequate.
Bromberg: Were you in touch with Hopkins?
Wolf: We were personal friends, actually. You know, there are two Hopkins’s. You must make sure not to mix them up. H.H. Hopkins in London, that's the one I'm talking about, and R.E. Hopkins in Rochester. They both will come in my story and I will distinguish them by their initials.
I was good friends with Hopkins, but the subject has been quite controversial. We had our disagreements but in a friendly sort of way.
Bromberg: It was controversial even in the fifties?
Wolf: Oh yes. That is, if you want to know more about the controversial part of it, I'll come to it later when we come to the conference — final, in fact, at the conference, we can finally review some of the controversies, by no means all of them. It has been a controversial subject since the early days of it. Even papers by Einstein and Lowett related to it about 1914, which then there was a lot of controversy, and we had our controversy with Hopkins too, to some extent in print, but of a different nature.
Hopkins's theory was very good when applied to practical problems of instrumental optics, but it didn't in my opinion touch on the basic studies, the (?) of it. That paper, I think was much superior. OK? But Hopkins's method is used quite extensively in instrumental optics, even now.
Bromberg: Were you already at this point, research problems were within the circle of your — astronomical research problems were —
Wolf: — to some extent, yes, because it soon became apparent, after I started working on it, that one of the interesting examples of coherence , really connected with astronomical problems, there is an old method of Michaelson for measure stellar diameters, an interferometric method, and soon after I started working on the subject it was obvious that one manifestation of coherence effects generated by essentially incoherent sources, stars , in the propagation process and how coherence is generated, and so my interest in astronomical applications in relation to it came up at the time.
Bromberg: Now, on the work of A.T. Forrester and his collaborators, already in ‘47 he has a letter out in the PHYSICAL REVIEW saying that he thinks that you can make beats between light beams.
Bromberg: Was that already attracting the interest of —?
Wolf: Now, of all the, what I would consider reasonably fundamental papers in the general area of coherence, not just theoretical but experimental, Forrester's paper and the ones, he wrote, I don't know if you have seen, Gudmundsen, one of his collaborators, and Johnson, I think, stand a little bit aside, I think partly because, in my opinion, they weren't written as clearly as many other papers, and in my opinion, other than papers, they would probably not have attracted very much attention at all if it were not for the Hanbery-Brown Twiss effect, which turned out to be relevant, also, the Hanbery-Brown Twiss effect because it generated a tremendous amount of interest, a lot of controversies, an d so it was soon found Forrester's experiments somehow were related, and they essentially showed very similar features essentially, possibility of correlation experiments from incoherent sources, but in my opinion they were never as clearly understood as the Hanbery-Brown Twiss papers.
Bromberg: Is that lack of clarity, these were electrical engineers.
Wolf: That's right. So was Hanbery-Brown, by training he is an electrical engineer.
Bromberg: I just wondered if it was because he did not have that kind of theoretical grasp of coherence.
Wolf: — yes, but you know, Hanbery-Brown and Twiss didn't have it either, really because it was at a time when the theory was still being developed, and there were just a few specialists who worried about it, but it was a more difficult experiment, and I personally, I really can't speak for other people, but I personally found it much harder to read than the Hanbery-Brown Twiss one, and I think, I may be wrong, but I think again to some extent, by the fact that the Hanbery-Brown Twiss papers had attracted so much attention and then people somehow realized that they were connected, and without Hanbery-Brown, it's not clear to me what an impact they would have made.
Bromberg: When personally did Forrester's paper cross your field of vision?
Wolf: Oh, not until the Rochester Conference in 1960.
Bromberg: I see, even later than —
Wolf: Yes. Even later. But I was much more aware of the Hanbery-Brown Twiss paper for many other reasons, not only because they were fascinating, but I was his colleague at Manchester University; after I left Edinburgh, I went to Manchester, and I had many extensive conversations with Hanbery-Brown, and we knew each other very well, and in fact I published a paper in the very early days of the Hanbery-Brown Twiss controversy on the subject. So I was, I followed it from the beginning.
Let me just correct myself. I'm not sure I became aware of it at the Rochester Conference. I might have heard about it already from Hanbery-Brown and Twiss in Manchester, but it just didn't make an impact on me. It was not until the Rochester Conference that I realized there was some connection. But it didn't make an impact on me.
Bromberg: Because he was one of the speakers and so you must have invited him.
Wolf: That's right. That's right. But of course, we had advisors. I was not the only person on the conference. But it didn't make any impact on me until then.
Bromberg: OK, now, the other question I asked you was, how to understand the kind of studies you were dong vis-a-vis the kind of study Dicke did when he worked on super radiance and I.R. Senitzky was working on this kind of problem also.
Wolf: There was no real connection at all. The problems I was interested in involved basically to understand how the theory of interference of light and diffraction of light really should be properly described in the framework of classical optics. It was really classical optics. We were writing a book on classical optics.
Now, as to what Dicke did in super radiance, of course he was interested in 21 centimeters, in atomic systems, which is a different problem. I don't know whether you know it but it's another example of a paper which people found extremely difficult to read. People knew it was an important paper, but even now, in fact, when he gave a paper at the Rochester Conference he said that he's giving it partly because his work has been — you know, not many people understand what he was doing before. Everybody realized it was an interesting paper, but very few people could understand it. And people even now are still working on the basic aspects of super radiance without understanding it properly.
Senitzky worked a little bit in isolation, but it was in basically the same area essentially, diffraction problems, system of molecules and cavities and so on, but my own work was completely independent of this, in the sense that theirs and mine had very little in common because they were really quantum mechanical or semi classical — Senitzky to some extent semi classical studies — whereas mine was entirely on the classical theory. I did some work later on quantum theory but at that time it was all classical.
Bromberg: So in that case I should probably also understand that you weren't much occupied with quantum electronics as a whole. No. Not at all.
Bromberg: Even with the theoretical attempts like Shawllow and Townesto say —
Wolf: — that's right.
Bromberg: How did you learn about the laser? Do you have any remembrance of your reaction to it?
Wolf: Yes. It's very much tied up with this problem of organizing the conference. I don't know to what extent you want me to go over this.
Bromberg: Yes, that's an interesting —
Shall I mention it because it's very interesting, very intimately connected with these things. You see, after I left Edinburgh in 1954, (Baum?) retired in ‘53, I stayed another year in Edinburgh. He was still continuing writing the book but much of it had been done by correspondence or my visit to him or my visit to him, he retired to Germany, I went from England to Germany to see him.
I went to Manchester in 1954 on a fellowship, research fellowship. That's when I met Hanbery-Brown and got to know Twiss. He was in London but he used to come up to Manchester in connection with Hanberry-Brown, and my fellowship was running out, I think either it ran out in ‘57 or ‘58, and in fact my last year in Manchester I was supported by this organization, Cambridge, it was called, how was it called ? Cambridge Research Laboratory, the last year.
And I was looking for a university job, and it was very difficult in those days to find one, and I had had an offer from the other Hopkins, R.E. Hopkins, who was then director of the Institute of Optics at the University of Rochester, to join the faculty in Rochester, and Hopkins, soon after he made the offer inviting me, was in England and visited me in Manchester and we talked about it.
I later accepted that offer, but during the discussion, he told me one of the first things he would like me to do when I came to Rochester was to organize a conference on coherence. He knew of my interest in coherence and of my papers and it was sort of my first assignment, and he wanted me, it was clear that I wouldn't get to the States much before the end of ‘58 or ‘59 and he wanted already to have the Conference on Coherence in 1959. I thought it was premature because I was just coming to the States and I didn't know how things were done in the States. I had to organize another conference in Europe before that, in Manchester, actually, on astronomical (?) optics, but I felt I knew nothing about how to do things in the States, the financial support and so on, so it was agreed it would be 1960.
So we started organizing this conference for 1960. I had been checking before the start actually some details about it, and what amazed me, when I came to the States, was how easy it was to get money for it. The Air Force Office of Scientific Research was waiting to give the money for this conference. We got it in a few weeks. Then the Optical Society of America was quite interested toco-sponsor which they did, and then, and University of Rochester, and we quickly set up a conference. I don't know whether you want me to enumerate a list of names, you probably have them, of people who were — yes, who were organizing this — and it was at that time that I realized that part of the interest in why we should organize a conference was that people were working on the laser, they realized that coherent light was absolutely crucial to it, and there were very few people who understood coherence theory. That there was a big gap still between the cluster theory of coherence and what would have to be done completely with the laser to understand its full status, apparently was a separate thing. Now, obviously they were closely related, and so that was the origin, how I suddenly became aware of the great deal of effort which had been put in provided, you know, essentially a resonant cavity with you know stimulated emission of radiation which we eventually did with the laser.
So I became really aware of this when we came to the United States in "59. But it was very quick, it was only a few weeks I needed to realize why Hopkins was so interested that Rochester should organize the conference and so on, and it just was very very easy. Money came easily. People were interested in, no problem.
So that was really the origin of it, but one thing which I noticed which is really quite an amazing coincidence , the following Our conference took place on June 27, to 29th of 1960. The announcement of the first laser by Maimen was in TIMES on July 7, 1960. Nine days after the termination of our conference. It was a strange coincidence.
Now, some people have told me that they heard from Malcolm Stitch (?) actually at the conference —
I heard that rumor recently too. I think that is correct. I wasn't aware of it then but since then people have told me of all rumors at the conference about it. But it is interesting in retrospect to now look at the papers which were presented there.
And how much of it was concerned with one or other aspect of the laser. There are those four examples, a paper by Snitzer there, on coherence properties of light in directed wave guide modes. And it was clear, if you read now, that he was very much interested in making a resonant cavity of the thing.
Of course there was the paper by already mentioned. There was an introductory paper by Fano which was very interesting, and stressed the confusion surrounding the whole subject of coherence, and he gave a number of examples, historic examples of people meaning different things by the concept and the amount of trouble it caused. In a similar sort of way, Purcell also talked who was chairman of one of the sessions, you know, Purcell from Harvard. Incidentally, Purcell was the first person who really gave a simple explanation of the Hanbery-Brown Twiss effect, you probably know about this, after all the controversy, his article in NATURE just put the subject straight, you know. He also referred to some of these controversies, and Fano mentioned how different people seem to understand completely different things by the concept of coherence, it was part of his study of the subject, but if I may just quickly mention one or two, there was a conflict (?) around 1940 on the one hand between Einstein and Hopf , and on the other hand between von Laue , about correlations and fluctuations in black body radiation. There were controversies about three or four issues of, the controversy between Einstein and Laue, Hopf's name appeared only actually on the first paper of Einstein, then it was Einstein and Laue, and I remember one paper said something like this. Einstein starts by saying, ""If,'' it's in German of course but the translation is something like this. "" If even an expert like von Laue doesn't understand this point, it's obvious it's a tricky subject which has to be explained more clearly.''
It was on the first controversy. It was by the way not completely settled until about the fifties by mathematicians when they properly described the stationary random processes. I could give you the references. It was a bit premature, the controversy, because the mathematics wasn't properly formulated by that time.
Then later there was a controversy between von Cittert and Zernicke, both of which made very important contributions to the classical theory of coherence. Von Cittert published a paper a few years later, I can find it, 1934, and Zernicke was either 1938 or "39, and as soon as Zernicke published his paper, von Cittert attacked him, that what he did before covers and subtleties, von Cittert wasn't right in that argument, Zernicke was actually, but that's another example of controversy.
And then later there was the controversy to some extent, it was not a very bitter one, between H.H. Hopkins and me, and of course there was later the controversy between Glauber on the one hand and Sudarshan , Mandel and myself on the other. The subject is full of controversies, and some of it came purely from language, not all of them, but the word coherence means so many different things to different people, and this was the sort of thing which Fano and to some extent Purcell were trying to straighten out at the conference in their talks.
Bromberg: Now, do you remember anything of the ambiance of that conference, or what excited whom, or any of these things that you don't get from the Proceedings, that we're only going to be able to get from memories?
Wolf: I think the paper, the introductory remarks by Fano, and the paper by Purcell were regarded almost as highlights, apart from the paper by Hanbery-Brown. Fano I think largely appealed to people who treat these problems in quantum mechanical language. One of the interesting things about the conference was that it was actually roughly an equal mix of classical optics people and people who eventually would be called quantum opticists, they used quantum mechanical methods. It's roughly equally balanced, and I look now at the abstracts, it's about half and half.
The people that I call quantum people were all very excited about Fano. He started from the beginning to treat the whole thing quantum mechanically, and then that was his prescription, how to do it.
Plus then of course there was his special way of real clarity of explanation of things, and that’s probably why the Hanbery-Brown Twiss effect controversy was eventually settled very quickly after his little paper appeared, in NATURE, that caused a lot of excitement.
I am now sure, Townes' name was in the program but he couldn't come to the meeting and the paper was presented by (Oliver) Heavens, and it was a paper on coherent and stimulated emission devices. I don't remember the details of that paper now, but reading it now I'm sure it must have been one of the central papers, because people were already aware, at least the people who worked on going from the maser to the laser realized that that sort of thing was a very basic thing.
These I would say were the main papers, plus Forrester's? paper of course, mixing light from incoherent sources. But again I think the number of people who understood Forrester's paper were very very small. I'm sure Fano understood it well, I'm sure Purcell understood it well, but there are subtleties in it which are not really made quite clear.
Now, these were the, I would say, the highlights of it, but there were other interesting papers there. Mandel's paper on superposition of light from independent sources was a beautiful paper, connecting, going from the ordinary coherence effects to transient coherence effects, and as you know since then Mandel has made very substantial contribution to this whole field. These were the early papers of his on the subject.
There were reviews of the classical theory of coherence presented by myself, also by Gamo, and summaries were made of what was known up to that time of the classical theory of coherence, and I look now at the review by, which appeared in PHYSICS TODAY by O'Neill and Bradley, as you know, they discussed this in some detail and summarized some of the things, so I think that makes possibly some contribution too.
There were practical things connected to ordinary optics, there were papers by Marechal and some of the well-known optics people, by Toraldodi Francia, — Toraldo di Francia's paper in retrospect, it's interesting too, but at that time probably nobody appreciated it. He had a new way of looking at the Smith Purcell effect. Now, this is an effect which you get when a beam of electrons passes close to a grating, that produces, and that produces, gives rise to radiation, and that was the forerunner of the free electron laser. Now, the way Toraldo described it, he had a new way of interpreting it, which was a very significant
step to understanding some of these types of interaction problems. But I am not sure now whether at that time it made any impression. A few years later I became very well aware of it, what an important paper it was, and as I say it was a forerunner of the free electron laser, so —
But very largely, if you want to know the highlights, I would say probably Fano's paper, probably Purcell's, although Purcell's was not a formal paper; Purcell just made introductory remarks as the chairman in a session. They were just about the main highlights, and I think, but I cannot be sure now, that the paper by Townes was given by Heavens.
Bromberg: Now, how did you personally get involved in thinking about the laser's coherence properties? Immediately after Maimen or what?
Wolf: Oh, once the news was out, about the laser, everybody in optics was thinking about the laser. I started asking a lot of questions about the coherence properties, and I think to this day some of them have not been properly qualified, even very excellent papers by very well-known people, the direction of it of the laser has entirely been attributed to the complete coherence, spacial coherence of the source. Now, since then, I don't want to overdo my own part in the subject, but since then, my group has shown that you can have sources with a very low degree of coherence, but if you design them properly they will produce even just as direction as a laser, the only thing is they will not be so intense. So it's clear that there are aspects of coherence and directionality which are not understood until very recently.
Now, in the last six months, in a paper with my former students
I got about looking at precisely the problem of coherence properties of laser modes. The theory of laser modes, which is largely based on a classic paper by Fox and Lee, is essentially a theory of monochromatic light. Monochromatic light is bouncing between two mirrors,
That approach, important as it is to the theory of modes, cannot explain anything about coherence properties of the modes because it's monochromatic. If you want to learn about coherence properties, you have to go to statistical descriptions. And I made an early attempt at this at the Quantum Mechanics Conference in Paris, I don't know the year now, "66 or "63, I had the forerunner to this, and the paper has been, actually, some of the people followed it up on some of the properties of coherence, some of the aspects of coherence modes, but recently we understood the whole thing much much better and we have a paper on the way, on this, but it's still far from the whole story. There are aspects of coherence with lasers which are entirely in its infancy.
It's a typical thing, especially when one thinks of applications. People think about what they can do with the laser, make it work, make it apply to something, devices and so on, but the basic problems, there are still quite fundamental problems about coherence properties of laser modes which are not understood.
Bromberg: It may well have a lot to do with the way in which research is financed and development, in this country, because there's so much pressure.
Wolf: That's right. That's right. It's very hard to devote one's time to the basic problems. There's no money for that.
Actually, for funding. It's very difficult. I have my own problems with it at the moment. But nevertheless these problems will eventually have to be settled.
Bromberg: Now, there seems to be , one of the things I'd like to get at is the way in which people from various specialties come together in these things.
Bromberg: And I'd like to try to get at a little bit the way in which optical scientists like yourself come into this whole field. I mean, between the Paris Conference of "63 and the conferences of ‘59 and ‘61 in that series, there's a real leap, because suddenly you appear and Glauber and Mandel, who weren't there at all, and that's the kind of thing I'd like to sort of explore a little bit, how —
Wolf: How it happened then.
Wolf: Well, I think part of the explanation of this is that laser physics started out to be an extremely rich field. You know, the seed for all of it is contained in Einstein 1917 paper, as you know. He was not asking the questions that more practically oriented people are interested in, but his whole analysis of the stimulated emission and the questions as you know played a very very significant role later in the whole theory of lasers. The subject has so many of these fundamental questions there and so many potential applications, it actually turned out that anybody reasonably competent, either in optics or electronics, could find a good problem in this. My own part, first of all, was the coherence aspect of it, but there are others — problems of diffraction and problems of interference of the light, and in the more recent years, for example, applications like optical conjugations, which brings in scattering theory, and again, a completely different type of people come to look at scattering of phase conjugated waves generated in four way ? mixing, than people who make actual devices with it.
It's such a rich field that a lot of people who are reasonably competent in some area of optics or electronics I think can immediately find something in it which would interest them and on which they could work, and I think this is just the basis of it.
Of course the situation with Mandel is a little bit different. Mandel understood the theory of photoelectric effects of light fluctuations much earlier than anybody else. He was stimulated by this little note of Purcell. Mandel immediately realized how important Purcell's approaches were and developed a very beautiful theory. I had known, I knew Mandel already in England when I was there, and of course that was one of the first things, when I came to the States and was told I should try to help to build up an optics group in the physics department, the first person I wanted to bring over was Mandel. It was obvious to me that he understood this aspect much better than anybody else. I knew that Purcell thought that that type of approach would be absolutely fundamental to elucidate statistical properties of lasers, and Purcell was absolutely right. It was Mandel's type of approach which followed on the Hanbury-Brown Twiss experiment and explanation by Purcell that just opened up the whole field.
So I was very anxious to get Mandel to Rochester, and we got him there eventually. As you know he's been there ever since.
Bromberg: I see, he came for a year, first?
Wolf: He came twice, I think, on two visiting appointments. One was a year. One was just a few months. He came first to our conference in 1960, and gradually we persuaded him to come over.
Now, I forget the beginning of this question, but this is a clear example of somebody from a really different field, somebody who'd had a lot of experience with electronics, who suddenly realized that a new era was opening up, and had the foresight to see it, who jumped into it, and he's done with optics absolutely the limits of what a fellow can do with the subject, and now it's a very fundamental matter. All the statistical properties of light are analyzed by photoelectric methods, and the whole analysis of basic concepts is due to Mandel, going back of course to Hanbury-Brown, the Hanbury-Brown Twiss effect, and just, you know, a little bit to it, which is a little bit controversial, it's collectively the later controversies with Glaubser and so on, Mandel produced a formula relating to probably distribution of photo counts, the counts of photo electrons as they were coming out of the photo detectors, with the statistical properties of the light. Now, his formulation was classical and semi classical. It was attacked later by a large number of people. I don't want to mention names. There were three or four people in print claiming, this is not the way to do it. Since at time there must have been dozens of papers and the final answer is, it works. Quantum mechanical formulation, rigorous quantum mechanical formulation, eventually gave the same formula. What it shows is that Mandel has extremely good intuition, and it's always easier to build on somebody else's work or read it in a different language, try to improve it, and one doesn't want to, one really doesn't do the calculations any better, but in a way, to me, in a relatively … he produced something which stood the test of time until now. It might have been not logically as sound as one would like if you could have produced these things quantum mechanically from the beginning, but the main thing is to get first the results and then worry about the details. So in my opinion this was one of the major contributions, and it follows very much of course Hanbury-Brown and Twiss.
Bromberg: Now, I think we have time for one more, one quite different thing I want to ask you about. At some point around 1961, two programs were presented to the Optical Society of America, a Rochester program and a Boston program. People were worried about whether there were enough optical scientists to handle the new work that was —
Bromberg: Is that something that comes out of your group or is that something quite different?
It is different. Let me explain that I originally came to Rochester to work in the Institute of Optics, but the Institute of Optics at that time was quite an independent body. It was part of the University of Rochester, but it was not attached to any college. It was a separate institute. Then about a year or two after I came to Rochester, there were some administrative problems at the Institute, and it was decided that the Institute would become part of the College of Engineering. By that time, by the way, I had a joint appointment also with the physics department but my main appointment was in the Institute of Optics. I was not the only one who had a joint appointment, there were several others.
At that point, when the Institute was being moved to make part of the College of Engineering, those of us who had physics affiliation were asked what did we want to do, did we move to the College of Engineering or did we want to go to the physics department in College of Arts and Sciences(?).
I had opted for Arts and Sciences. So did many others. But that didn't mean we didn't like the Institute and collaborate with the people, as you know; now I have again a joint appointment with both of them, the main one being the physics department.
But the Institute of Optics was developing its own program for broad linear optics because it was obvious one needed to go beyond the traditional subject which people were trained in optics.
There was a group in Boston, largely around Professor O'Neill who has also input on what to do about optics, but I think many of these programs — you know, it was just in thin air. There was not so much done, certainly not in the Boston group, and not all that much done in the Institute of Optics really, to meet all the needs which were then, you know, came up.
I don't know at what point the Institute of Optics, the Optical Sciences Center in Arizona was. Do you happen to know?
Bromberg: Not really.
Wolf: But I think it was around that time. There was a need to get more optics, and it went very very slowly, at Rochester. In fact, if anything it was to some extent the physics department that started some of the more modern things, by bringing Mandel in, and another young man later, he's now a full professor, (J.H.) Eberly and myself, who were more in, what you might call perhaps the more modern optical physicists.
I still largely was working with classical optical physics at that point; quantum optics was largely done by Mandel and Eberly. I have written quite a number of papers in quantum optics but my emphasis still was in classical optics. But in my opinion... (off tape)