Peter Roll is retired, after 25 years as a university administrator of technology. He is currently working on the development of a community web portal for the retirement community in which he and his wife live, near Austin, Texas.
My perspective on the 1965 discovery of the CMBR is quite different from that of other contributors to these essays. Dave Wilkinson and I had our first measurements of the CMBR in the summer of 1965. We satisfied ourselves and our colleagues - Bob Dicke and Jim Peebles - that they were valid. Shortly after this, I left Princeton to join the staff of the Commission on College Physics in Ann Arbor. One thing led to another in my career, and by 1971 I had gone into academic administration full time at the University of Minnesota. Research became, for the remainder of my life, a spectator sport in which I played a support role in a variety of administrative ways. I've remained an active spectator, keeping up with scientific press reports on developments in which most other authors of these essays were directly involved.
Dave Wilkinson and I began work on the Princeton measurement of CMBR in 1964 - I had finished work on the Eotvos-Dicke experiment the previous year and we had written it up for publication (Roll, Krotkov and Dicke 1964), and Dave Wilkinson had recently joined Bob Dicke's research group. Dicke set both the theoretical context for our work - looking for remnant radiation from the big bang - and the experimental approach -using the Dicke radiometer he had invented in 1946 at the MIT Radiation Laboratory. Jim Peebles was doing the theoretical calculations, keeping us informed of how they were related to our experimental work. Dave and I were experimental physicists with no previous experience in radio astronomy and little experience working with microwave electronics and liquid helium. But we learned, and we designed and tested the equipment, with encouragement from Dicke and other groups in the Palmer Physical Laboratory. The work progressed well, and by February 1965 we expected to get data that summer.
In his last interview, Dave Wilkinson described the telephone call from Arno Penzias to Bob Dicke during one of our weekly lunch meetings. Dave's description is exactly as I remember it, with the exception of the length of the call. Dave described it as short, about 5 minutes, while I remember it as long, about 30-40 minutes. Visits were exchanged with Penzias and Bob Wilson at the Bell Laboratories Holmdel site and Princeton, and we all knew that the Princeton group was going to be number 2 rather than number 1 on the discovery.
What did we discuss during these exchange visits? Two things have stuck in my mind, though I can't trust my recall too far after 40 years. The first is that Dave and I quizzed Penzias and Wilson on the details of their equipment - how they had dealt with the many difficult problems to eliminate sources of systematic errors. They satisfied us that they had done this properly, and we shared information on what we were doing about these same problems. We learned about pigeon droppings in the horn-reflector antenna that Penzias and Wilson were using. Our equipment at Princeton was smaller for the shorter wavelength (our 3 cm versus their 7.3 cm) and we could more easily cover it when not in use. Even though the birds themselves were plentiful on our Guyot Hall observing tower (Figure 4.22), heat radiation from pigeon droppings in the antenna was not a significant issue for us.
The second detail we asked about was what Penzias and Wilson were looking for when they started their measurements. My recall is that they
started out to make an absolute measurement of the radio flux from the Andromeda galaxy, for which they would need an accurate measurement of any background flux from the sky around Andromeda. I suspect my recall of this detail may be, at best, a little oversimplified and incomplete.
The story of how the two research groups learned of each other is told well enough in the 1978 Public Broadcasting System Nova program, Whispers from Space. Jim Peebles gave a talk on his work at the Johns Hopkins Applied Physics Laboratory in early February 1965, with Dicke thinking that we were far enough along with our apparatus to finish and get data before anyone else could do so - because it would take them longer than that to build the equipment. Professor Bernard Burke, a radio astronomer at MIT, either attended the lecture or heard about it - he knew a little about what Penzias and Wilson were up to and suggested to them that they should contact Dicke. What Bob Dicke quickly recognized was that, if someone already had the apparatus and had started or completed the measurement, they would beat us.
We had been pretty sure that, if what we were looking for turned out to be cosmological, it would be an important scientific discovery. Realizing we would be number 2 created, I think, a certain amount of "awkwardness" among those of us at Princeton. It was a disappointment, of course. I hope I speak for Dave Wilkinson and Jim Peebles in saying that we all felt more disappointment for Bob Dicke than for ourselves - Bob had been so close to a big one more times than most scientists. Dicke, I suspect, felt more disappointment for Wilkinson and me than for himself.
We understood that these things happen, and that being number 2 was still very important. At that time, the explanation was by no means certain, and none of us could be dead sure that there wasn't something wrong with the measurements. Penzias and Wilson were initially stymied by what they had, and we hadn't yet gotten data to examine. Number 2 would be important to confirm the result and get a second point on the spectrum - if it wasn't thermal, then our explanation could not be correct. At Princeton, we all got back to work, knowing that we were on to something important.
Shortly after the exchange of visits between the Princeton and Bell Labs groups, Bob Dicke informed us that each group would publish a letter, to appear back-to-back in the July 1965 issue of The Astrophysical Journal. The first letter would be by Penzias and Wilson (1965a) announcing the discovery, followed by a second from the Princeton group interpreting the result as remnant thermal radiation from the big bang (Dicke et al. 1965). We got to work on our letter, based largely on the work of Peebles and Dicke, but including a description of the work Wilkinson and I had begun.
Over the years, the "awkwardness" associated with the discovery has been, for me, explaining the situation to others. Broadcast of the PBS Nova program in 1978, followed shortly by the Nobel Prize award in the fall of that year, was the first time this work had been in the public eye enough to trigger these questions. Did I - or the Princeton group - feel in some way "cheated by circumstances?" I certainly didn't feel that way. All of us with the Princeton group at the time were sorry that Bob Dicke had not been included. I think we all understood also, in different ways, why the Nobel Committee did not do it that way. A 25-year history of research findings and conjecture preceded the discovery - it is summarized well by Peebles and Partridge, and documented in detail by many of the other essays in this collection. If nothing else, this left a confused situation for the Nobel Committee, and one about which there are differences of opinion. The discovery of the CMBR certainly deserved recognition. The Nobel Committee made a good choice, and it may have been the only one they could make.
My family and I first watched the PBS Nova program as a rerun in January 1979, after its first broadcast a few months earlier and after the Nobel Prize had been announced. We watched it, in fact, in an empty hospital room that nurses had set up for us across the hall from where my wife was recovering from surgery. Our children ranged from 9th grade to college junior in age, and I had previously told them the story and explained why Dave Wilkinson and I should not have received or shared the Nobel Prize, I thought they understood. But when Penzias and Wilson appeared on screen showing their apparatus, the kids began exclaiming, "Is that them, dad - are those the guys that won the prize? ... Boo! Hiss! Boo! You guys took the prize away from our dad!" So I explained it to them again. Their outburst on that occasion was a somewhat more candid and immature way of expressing what many others have asked. With a few more years behind them, I know they understand now. Soon I will show the tape and explain it to two very bright grandchildren.
I've told the story of this discovery many times, including to classes on cosmology I've given several times in the Senior University of Georgetown. (I rely on the 1978 Nova video tape to tell the part about Princeton and Bell Labs, however, and I leave the rest to questions.) This Senior University is an "institution" formed by several fellow residents of the "active adult community" in which my wife and I have lived for the past ten years. Its 600-odd students are almost all nonscientists - bright, mature adults with a lot of experience and accomplishments in their lives. They are fascinated not only by the story of how our universe began, but also by how and why scientists do this kind of work and arrive at some really strange conclusions - conclusions that are supported by a web of evidence from many different fields of research. Despite the fact that they are supposed to be objective, each scientist experiences the story personally and tells it differently. In this regard, science is no different than any other area of human endeavor.
The inside story of the discovery of CMBR, and the understanding of our universe to which it led over the past 40 years, is a magnificent example of the scientific method - messy, as it really is:
Looking back from 1965: Early research in the creation of heavy elements, intergalactic molecular spectra, the radiometer developed in radar research at MIT - a quarter century of missed hints and clues, almost but not quite pieced together more than once - finally pieced together, and two groups coming up with results at nearly the same time.
Looking forward from 1965: A discovery that was initially controversial has been so well documented with a variety of measuring techniques, and new and much more sensitive detectors used at high altitudes, from satellites, and over a wide range of wavelengths. The CMBR was first examined and thought of as uniform in all directions. It became possible to measure the direction and speed of our Galaxy's motion through absolute space by looking at a small asymmetry in the intensity of CMBR - hotter in one direction than in the opposite direction. Theories emerged on the earliest history of the universe, including Guth's strange superexpansion in the first instant of the big bang; and on how and when stars and galaxies began to form. These theories had to be consistent with one another, or they wouldn't be accepted. It became possible to calculate the distribution of tiny fluctuations in the CMBR and to measure these fluctuations from the WMAP satellite, distinguishing between some valid and invalid theoretical concepts and establishing numerical values for some of the important properties of the universe in which we live.
What a different understanding of the universe this is now, compared to the time when the steady state and big bang theories were actively contending with one another! During these past 40 years, many other concepts and variations were tried and found wanting, either because of theoretical inconsistencies or observations that did not support them.
The story of this one discovery has several of hallmarks of the scientific method, in addition to the messiness mentioned above: Hypothesis about what you are looking for. How do you expect the research to turn out? Whether this concept is well founded or speculative is beside the point. Dicke's real contribution to the original CMBR work was just that - it was his idea to look for red-shifted thermal radiation from the big bang. A corollary to this principle, however, is to be skeptical and challenge your own conclusions, especially if they support your biases. In drafting the second of the back-to-back letters to The Astrophysical Journal in the spring of 1965, Dicke incorporated a statement that the CMBR detection was evidence for a closed universe that would one day contract back on itself - a concept tied to his work on the Brans-Dicke scalar theory of gravitation. In a chance meeting of the two of us, I argued that he should remove this statement, leaving intact the discussion of ramifications of the CMBR for open and closed models of the universe (flat was thought to be very improbable at the time). I was quite uncomfortable disagreeing with a person for whom I had the utmost respect. Neither of us could have guessed that results, 40 years later from the WMAP satellite, would show the distribution of tiny fluctuations in the CMBR and confirm a flat universe so convincingly.
Careful documentation. When Dave Wilkinson and I completed our first measurements, we took time and care to document, in a 1967 article in the Annals of Physics, what we did and how we did it, in complete detail (Roll and Wilkinson 1967). I had done this earlier with the results of the Eotvos-Dicke experiment, because the validity and limits set by a null result are the important part of the experiment. We did likewise with our CMBR measurement, because, at the time, it was controversial and not at all accepted that it was a thermal spectrum; if there was anything wrong with our methods and analysis, we wanted others to be able to find it. In hindsight, this was completely unnecessary. There have been so many measurements by so many techniques confirming the blackbody properties of the CMBR and more, that the specifics of how we did the second measurement have become almost irrelevant. Nevertheless, it was important to both of us at the time to complete the job properly.
Persistence. There are three examples of this among the people I worked closely with at Princeton. The first is Bob Dicke, who devoted the last half of his professional life to gravitation and cosmology - devising conceptual/theoretical models, experiments, and observations to understand better the nature of this basic law of physics and the physical nature of our universe.
The second is Dave Wilkinson, who spent his entire career after graduate school following the trail of the cosmic microwave background, eventually into space. Dave's scientific legacy is his two decades of work on satellite observations of the CMBR, culminating in the Wilkinson Microwave
Anisotropy Probe satellite. Results are still coming out of data from WMAP, as recently as two weeks before I write these words. All of us who knew him, even from way back, grieve that he is not still among us to witness the results of his dedication.
The third is Jim Peebles, who started as Bob Dicke's student and stuck with his study of galaxies, cosmology, and related matters from a more theoretical perspective - but still related closely to observations and measurements. I am not as familiar with the details of Jim's work since I left the field, but I hear about it often enough to know that he has been at it consistently and persistently for 40 years.
I'm quite sure that all three of my former colleagues have contributed as much, to science and society, by the students they trained and mentored as by the research they have pursued. Some of them have become successful scientists in their own right - others have gone off into other fields, as I did, and contributed in other ways.
An autobiographical appendix: notes on what drove me to physics, and then to leave for a different career. I came to physics from a family with no particular interests or talents in things scientific. From an early age, I had a knack and interest in things mechanical and quantitative. I entered Yale as an undergraduate with many interests. Before my senior year, I had taken no physics at Yale beyond a noncalculus introductory course. (I did sit through several graduate courses at Heidelberg during my junior year on an exchange scholarship.) My first job out of Yale (nuclear reactor design at Westinghouse) and my graduate work in experimental nuclear physics back at Yale were both interesting and rewarding. But I've also played the French horn all my life. My motivation for physics was at least partly to understand the physics of that treacherous instrument, so that I might improve my skills as a performer. This, however, was not a fashionable area of physics research, and gravitation and cosmology turned out to be far more interesting.
Finally, when I became a full-time administrator at the University of Minnesota in the 1970s, I was able to continue teaching a course in Musical Acoustics and engage in a little research and dissertation supervision with the Departments of Music and Music Education. I learned from the late Arthur Benade (Case Western Reserve) that the basic physics of the French horn and other brass instruments is governed by the Webster horn equation (Bell Labs, ca. 1916), which is none other than the Schrodinger equation with a transformation of variables. And I did learn how to play the horn better because of this work in the 1970s.
In 1965 I left Princeton for a year with the Commission on College Physics in Ann Arbor. A major activity that year was a report on "Computers in Physics Education" - the first ever report on the role of computers in higher education. When I joined the Physics faculty at the University of Minnesota in 1966, I was quickly identified as "... an expert on computers in education ..." By 1971 I was serving on so many committees, doing interesting work for the University and the state, that I moved full time into academic administration, with a portfolio including computers; radio, television, and audio-visual services; and library technology. In 1984, I moved to Northwestern University as Vice President for Information Technology, leaving behind my vestigial teaching and research in musical acoustics. From there I moved in 1992 to Executive Director of netILLINOIS, a nonprofit internet service provider mostly for Illinois educational institutions in the early days of the Internet. In 1995, I retired and moved with my wife to a new Sun City development in Georgetown, Texas.
In hindsight, it turns out that the theme in my life since 1971 has been networking and communities, rather than physics. This began with my appointment to a Cable Television Advisory Committee of the Metropolitan Council of the Twin Cities in late 1971, where the theme was cable TV as a community service network. Through most of the 1970s and 1980s, I was a board member of EDUCOM, an organization that pioneered networking to support academic communities and introduced higher education to the Internet. At Northwestern, I set the stage for a proper networked campus, though it did not get far off the ground during my tenure there. As I approached retirement, it was clear that the Internet was the platform for the "community network" that so many of the activist younger generation were promoting in the 1970s in Minnesota. And so I moved to Sun City, Texas, with an interest in seeing how the Internet might become a community network as it matured. And this is a work in progress. We started with a Computer Club that now has 2000 members (out of 7800 residents) and are finally in the process of implementing a community web portal, which will be our community network.
Throughout this 35-year period, the scientific and engineering research communities have been the creators of the platform for community networks of all kinds - ARPANet, BitNet, Usenet, TCP/IP, and all the others. These networks migrated into the larger society, finally, after 1989, when Tim Berners-Lee developed the World Wide Web at CERN, and in 1993 when Larry Smarr, an astrophysicist and Director of the National Center for Supercomputing Applications at the University of Illinois Urbana-Champaign, fathered the first web browser, Mosaic. The Internet as we know it today was catapulted into society and the economy by the particle physics and astrophysics research communities, as a tool that has improved scientific communication and made progress in science faster, more efficient, and more accessible. It has transformed not only research, but also society and the economy. Even retirement communities such as the one in which we now live.
One of the issues which interests many of our fellow retirees is why the US taxpayer should fund research in basic science. Cosmology really doesn't have that much impact on everyday life. I conclude my Senior University classes in Cosmology with this question: What is the return on this investment in basic research? The answer to this is now unbelievably easy. The economic impact of the Internet is the return on investment in particle physics and astrophysics research for the last n years - you pick the number of years, and the dollars work out just fine.
But this economic impact is all an accident - it's not why any of us do or have done research in things like the CMBR - those reasons are much more personal and complex.
Was this article helpful?