This year’s Nobel Prize in Physics was, in a way, a foregone conclusion. The 1998 discovery by two teams of scientists that the expansion of the universe is accelerating—under the influence of something that scientists have shruggingly come to call dark energy, which later studies have revealed to comprise 72.8 percent of the universe—was one that everybody assumed would win the Prize. It was only a matter of when.
Only somewhat less foregone was who would receive the Prize for the discovery. The leaders of the two discovery teams—Saul Perlmutter, of the Lawrence Berkeley National Laboratory, and Brian Schmidt, of the Australian National University Mount Stromlo Observatory—were shoo-ins. As the lead author on Schmidt’s team’s discovery paper, Adam Riess, a postdoc at the University of California, Berkeley, at the time of the discovery (now an astronomer at Johns Hopkins University), stood perhaps—perhaps—slightly less of a chance, but the difference was in the nature of a four- versus a five-sigma result.
Even the allocation of the award was more or less foregone. Perlmutter would get half of the 10 million Swedish kroner ($1.44 million) prize, and either Schmidt would get the other half or he would split it with Riess. The latter scenario—50/25/25—is indeed what the Solomons of Stockholm decreed, and this past Saturday the three new laureates received their medals from His Majesty King Carl XVI Gustaf of Sweden.
Yet in the two months since the Royal Swedish Academy of Sciences announced the Prize, on October 4, many members of the discovery teams have found themselves experiencing what one astronomer described to me, via e-mail, as “a bag of mixed emotions.”
The problem isn’t that the wrong persons won. The problem is that the right persons didn’t.
According to Nobel bylaws, the Prize in Physics (as well as the other categories) can go, at most, to three living persons. That rule might have made sense in 1901, the year the first Prizes were awarded. But back then, science was still in the workbench (or, for Einstein, the patent clerk’s podium) era: one mind, one mission, one eureka moment.
That culture began to change when the UC, Berkeley, physicist Ernest Lawrence invented the “proton merry-go-round,” or what we would call the cyclotron. His first model, in the late 1920s, was five inches in diameter, but it served as the prototype for the particle accelerators that have come to symbolize Big Science, for better or worse.
Worse, astronomers would say. That Saul Perlmutter was at Berkeley Lab when he helped begin a collaboration among physicists that would discover evidence for dark energy is a coincidence. But the Big Science, top-down, the-same-lead-author-on-every-paper culture is precisely what the astronomers on Schmidt’s team tried not to emulate. They favored a more democratic, bottom-up approach: the guy (they were all guys) who did the most work on a paper got to be lead author.
Either approach, however, was going to involve more than one or two people per team. By the 1990s, astronomy had outgrown its own workbench era—the workbench being the telescope on a remote mountaintop, where astronomers would observe whatever facet of the heavens interested them before popping into the accompanying darkroom to develop photographic plates. Astronomers today aren’t generalists. They can’t be. The diversity of the science and the complications of technology have forced the field into greater and greater levels of specialization.
In the case of the two discovery teams, they were working not just with stars but with exploding stars, and not just with exploding stars but with a particular species (Type Ia). They needed spectroscopists, photometrists, coders. They needed experts in metallicity, redshift, and red dust. They needed designers and engineers to exploit existing technology and to design new instruments. They needed gruntwork from grad students.
They needed teamwork.
“The group has discovered that in all the hoopla, there has been no indication that it was actually a team that did the work,” Nicholas Suntzeff, now at Texas A&M (and a guest contributor to LWON), wrote me two days after the Nobel announcement. Alex Filippenko, an astronomer at UC Berkeley both then (1998) and now (and the subject of a recent LWON post), wrote in an email to friends and colleagues after the Nobel announcement, “All together, the two teams had 51 scientists, each of whom contributed significantly (and in some cases, a large amount) to the research.” The astronomers could take solace that at least members of the community would know how collaborations work. But the public? Posterity? As Britain’s Astronomer Royal, Martin Rees, said, “It would have been fairer, and would send a less distorted message about how this kind of science is actually done, if the award had been made collectively to all members of the two groups.”
Nobody was expecting the Nobel Foundation to agree to change the rule. But as the members of the two teams took to the Internet and old-fashioned phones to discuss the Prize, they found themselves trying to come to terms with the emotional fallout from the rule.
“If you are like me,” Suntzeff wrote to his team members the week of the announcement, “there is bittersweetness in this award. I am sure all of us agree that Brian and Adam deserve the award, and deserve to represent the Team’s success. But the reality is that there remains a huge difference between winning the Prize and not, which does not reflect the nature of our collaboration. Even though it is unfair that team effort will be forgotten quickly now by history, it will be the reality.”
Suntzeff had already experienced his own brush with obscurity. The original official Nobel press release said that Schmidt “organized” his team. Which he did—only he did so with Suntzeff. In 1994, during a discussion at the headquarters of the Cerro Tololo Inter-American Observatory in La Silla, Chile, where Suntzeff worked at the time, the two astronomers had agreed on the goals and allocated the responsibilities. Two days after the Prize announcement, Schmidt contacted the Academy, and as a result the Nobel website now reflects the facts behind the founding of the collaboration.
But even that collaboration grew out of a previous collaboration, the Calán/Tololo supernova survey, to which Suntzeff belonged. And the founder of that collaboration, the Chilean astronomer Mario Hamuy, had designed the survey specifically to serve as a first step toward the work that the Nobel committee had now immortalized.* After the Prize announcement, Hamuy went on Chilean television and, visibly upset, decried the lack of recognition for Chilean astronomy.
“I cherish the friendship of the group,” Suntzeff wrote to his collaborators. “I hope that our friendship will withstand the contradictory feelings of elation, jealousy, pride in accomplishment, and the sting of lack of recognition outside our community”—or what one collaborator called “feelings of disappointment, and disappointment with ourselves for being disappointed.”
“There is nothing wrong with these feelings,” Suntzeff concluded, “and we will live with them for a long time—they are natural. But just remember, we all played key roles in discovering three-quarters of the Universe, and no one else in history save our two groups will ever be able to say, ‘We discovered most of the Universe in 1998.'”
Then he went out and gave a talk to about a hundred Texas A&M faculty and students on “Almost Winning a Nobel Prize.”
For the record:
# # #
* Just prior to the publication of the discovery paper, Hamuy removed himself from the collaboration over a priority dispute with fellow team members; for the details, see pages 105-106 in my book on the discovery.
Credits: http://www.nobelprize.org/; detail from Laurie Anderson’s Big Science album artwork; http://iopscience.iop.org/
There is an interesting conversation on fb about this.
From David Arnett, National Academy member, Professor of Astronomy and Physics, University of Arizona, and pioneering theorist in supernova physics:
Dave Arnett:
It is probably a fallacy to think of prizes in science in a personally territorial way, as easy as that is to do. We should regard the prize winner as a representative of the scientific community in which the work as born and developed, and not the sole source. The task of a prize committee is to construct a plausible story about events they did not personally participate in, and the person associated with the best story gets the prize.
I replied:
quite true! The only problem with that is that it will introduce a bias favoring those in the US where the science stories are made. Ask a European or Canadian astronomer! Look at most of the quotes in Overbye’s articles in the NYT. They are almost always from US astronomers. We in Chile felt particularly left out of the story as it propagated in the physics (not astronomy) community. We were either not mentioned, or if so, usually it was as “and Chilean astronomers” as if to wave one’s hand at some nameless astronomers down there somewhere.
However, ultimately the Nobel Physics Committee did an excellent job of recognizing that the High-Z Team was truly a team and that we worked as a team. They have rarely done this before, and I appreciate their effort to write the story correctly.
Dave Arnett:
As the Stones sang, “You can’t always get what you want…”. Nevertheless, for an institution of mere primates, science does reasonably well, though it is not perfect. There WAS a Nobel for the supernova work, which I appreciate. Not ALL good people were rewarded, but SOME good people were. In my opinion, the prize should have included the team who discovered how to empirically calibrate SNIa’s. As for the regional press biases, they are awful sometimes. Unfortunately these biases extend far beyond the science news.
Thanks for forwarding, Nick. Great stuff.
And thanks for linking to the piece on fb.
Re Arnett: Not only do US science writers have a bias toward US scientists — I’ve been guilty of “Chilean astronomers” myself — but as Richard says, more and more astronomy is done by bigger and bigger teams, so finding the “person associated with the best story” is going to get crap-shooty.
Nice post, but I’m still wondering why, if dark energy was “inevitably” going to win someone a Nobel, dark matter wasn’t equally inevitable. Vera Rubin is still alive, after all. So, for all I know, is Kent Ford.
Thoughts?
Mike:
Dark energy was unambiguous. A “moment” of discovery, by two teams (despite the infighting over which actually got there first). It was a before-and-after discovery, and everybody involved knew in early 1998 that if the result held up, it was Nobel-worthy.
Dark matter was ambiguous in several ways. First, it was a discovery over time. Nobody said, “Aha! Dark matter!” They said, “Weird!” Then a few years later, “Still weird!” Then a few years later, “This weirdness isn’t going away….” Then a few years later, “Okay then–dark matter. Everybody on board?” Second, it was a discovery by several. The first Rubin/Ford was 1969/70, and a steady accumulation of papers by Rubin/Ford and *many* others (including Mort Roberts, working the radio part of the spectrum, and still with us), *plus* powerfully persuasive N-body simulations by Peebles, Ostriker (and Peebles, Ostriker, Yahil), finally led to consensus c. 1978-80. (See Chs. 2-3 in my book, if you want the details.) All that said, I think that Vera, Kent, and Mort should share the Prize. And I know Nick has thoughts on this subject, if he wants to rejoin the conversation here. But I don’t think a dark matter Nobel ever had that whiff of “inevitability.”
Thanks, Richard–I actually know the story from working on my own 1993 book, “The Light at the Edge of the Universe.” And I can see the argument that it somehow has to be a single “aha” discovery to qualify for the Nobel.
But that just adds another dimension to the questionable assumptions behind the awarding of Nobels. You rightly point out that the prize doesn’t really reflect the fact that modern physics often involves many crucial contributors beyond the allowed three. If the prize can only go for “aha” discoveries, even in the face of such an important phenomenon like dark matter, there’s something wrong with that as well.
I do agree with you about who should get the DM prize.
Interesting and thoughtful piece Richard, that to me brings into sharper focus a very important shift in the doing of science in astronomy. And that is the increasing organization of researchers into large teams in astronomy/cosmology. Big Science as this is referred to is not the only way to do science successfully. We ***absolutely need many intellectual cultures to survive, big teams, small teams and individual generalists.
Given that progress in our understanding requires creativity and innovation, we cannot afford to lose the individual workers who will take intellectual risks more readily than teams. I find it very sobering about how all of this is shifting,
and how the culture is changing rapidly and making it harder and harder for generalists to survive and succeed. One of the other perils of large teams is that it is by definition high resource requirement science and that is very much more at the mercy of the vagaries of federal funding and politics. Keeping many different styles alive, that require different levels of funding will prevent quenching of science due to the general economic crunch and the disruption of large science projects. Its just simply much much harder to be recognized if you are not part of a team and this will discourage younger people from getting into a science. What attracted me as a young girl into science was not hey let me part of a large team …..
In full disclosure, I am one of those generalists.
Mike: Thanks. But just to be clear, I’m not saying that a discovery has to be an “Aha” moment to win the Nobel. I’m just saying that an “Aha” discovery with unambiguous discoverers like dark energy has a much clearer path to Stockholm than a slow accretion discovery with multiple discoverers like dark matter. At this late date, dark matter is a longer and longer longshot, but if Vera, Kent, and Mort (or some combination) won, I wouldn’t fall off my chair. (Or at least any more often than I already do.)
It is my somewhat patchy memory back to graduate school in the 70s at Lick Observatory that remembers that dark matter was not a topic of proper colloquia until Rubin and Ford made their observations and wrote their papers. Most of the dark matter speculations came out in conference proceedings and not in the refereed journals. Vera told Richard and me it was impossible to discuss dark matter per se and get it by the referee in the 70s. But as you say there was no aha! moment. Sandy Faber came to Lick around 1974 and used the studies of velocity dispersions in elliptical galaxies that revealed *modest* amount of dark matter. At the same time the tidal radii of dwarf galaxies was also used to infer dark matter, and also the velocity dispersions of old stars in our Galaxy. In my opinion the discovery of the high velocities of stars in dwarf galaxies by Marc Aaronson around 1980 pushed the subject past a tipping point into acceptance. Marc tragically died while observing at Kitt Peak, and his legacy has been somewhat forgotten.
But the one constant in all this was the rotation curve of galaxies. This started the sea change in how we look at matter. And Rubin et al fought the establishment with their interpretation, which ultimately was adopted by astronomy. I see a strong parallel with the Chemistry Prize this year on quasicrystals. Here too the scientist fought for years, decades, to get his discovery accepted.
To me, it is clear that Rubin, Ford, and Roberts are clearly deserving of the discovery of dark matter.
Priya: Thanks for this thoughtful comment. An irony of the whole dark energy saga is that one of the discovery teams (High-z) was pushing back against the high-energy physics model of the other team (Supernova Cosmology Project), as I mention in my piece, yet that teams’ discovery is what has fostered an industry of dark energy research that requires, as you say, “high resource requirement science” that depends on “federal funding and politics”–aka, Big Science. Look at the fiasco (which I’m hoping isn’t too strong a word) surrounding JDEM/WFIRST, which would have tried to constrain models of dark energy (cosmological constant or quintessence, for those of you visiting this conversation who don’t know much about these issues). Simon White warned of what might happen if a generation of astrophysicists (and members of related fields) put all its energies into a single goal. From a sociology of science perspective, which we outsiders can afford to indulge, it’s a fascinating development; from the practitioner of science perspective, it must be a constant source of tension. Thanks for adding to the discussion.
The big vs individual science clash cuts both ways. As in the original essays, prizes and credit tend to go to the leaders of teams, not really recognizing the necessary efforts of team members. This is a problem in physics just as much or more than in the rare astronomy prizes. But also, because astronomy has until recently had a culture of individual investigators, astronomers haven’t built up ways of crediting the work of team members and builders. For every person who thinks they are being crowded out by large teams, there is someone inside a large team who thinks they can’t get credit for their work outside the team.
Particle physicists have sort of worked out apportioning credit, although I understand their internal politics are extremely complicated. Astronomy isn’t really anywhere near the particle physics scale yet; 25-50 people is a “large” observational collaboration.
I don’t agree that WFIRST is a fiasco or that it’s a good example of what we’re talking about here. WFIRST is falling victim to federal budget constraints, there’s nothing wrong with the project itself. Further, if WFIRST flew it would be an observatory, producing a survey that would be useful for many extragalactic science projects beyond measuring one or two dark energy numbers. (Probably also true for galactic science, but I can’t say as much about that.) I sort of agree with what Simon White was trying to say about general purpose observatories versus single-mission experiments, but some of his examples were not well chosen to mark this distinction.
Benjamin: Thanks for the comment. I just want to add that by “fiasco” I didn’t mean the science behind WFIRST (and before that JDEM, and before that ADEPT, and SnAP, and DESTINY), I meant the infighting between DOE and NASA, between NASA and ESA, between certain newly-minted Nobel laureates representing competing projects, all of which led to a scientifically sound, even ambitious (in the observatory, multi-disciplinary sense that you mention), project that was, given federal budget constraints and Decadal survey restrictions, never going to get off the ground in this decade, and now is as good as dead. I talked about this with Brian Schmidt in 2007, at a dark energy conference where just about every talk was haunted by Simon White’s then-recent paper, and he was worried that dark energy was pushing astronomy into new budgetary realms that were unsustainable on a cost-benefit basis. Hope this clarifies my intention.
Richard, thanks for the response. Here is a thought experiment. Suppose that the decadal survey had picked IXO, the next big X-ray mission, as its top priority instead of WFIRST. And then a year later the Federal budget and the JWST funding envelope had made it obvious that IXO could not happen in the next decade. Would we then call the IXO concept design process a failure or a fiasco? I don’t think so. What you saw was ugly, but it was also the result of seeing out in the open the arguing, self-dealing, back-and-forthing, occasional nastiness, inscrutable NASA/DOE/aerospace politics, and scientifically motivated legitimate compromises that go into all large experiments, as far as I can tell. Look at the twists and turns in the establishment of Kitt Peak, or in Rieke’s book on the building of Spitzer, for example. (“Last of the Great Observatories.” Disclosure: Rieke is my boss, but anyone who knows him will tell you he’d never pay me for mentioning his book.)
WFIRST, IXO, LISA, and a number of other future experiments that may or may not happen (including ground-based telescopes) share the problem Brian Schmidt saw. If you do a big mission/experiment, it is hard to figure out how to follow it up without building an even bigger and more costly experiment. At some point, does the cost go beyond what our society is willing to spend on astronomical curiosity? I think the main driver is how flush with cash the country is at any given moment.
What Simon talked about is a real worry for anyone trained in a classical astronomy mode. So far there are enough of those people influencing astronomy policy that most or all of the large DE experiments have also advertised that they will produce a dataset useful for many astronomical projects outside the main collaboration. This is a product of astronomy data culture, and is not the way particle physics experiments operate (for example, LIGO). Astronomers will attempt to keep this culture, but will they succeed?
Thanks, Benjamin. Well, I did say I hoped “fiasco” wasn’t “too strong a word”! I’ll certainly go along with you on “ugly,” though. You’ve made excellent points, and thanks again.
Benjamin, you have captured exactly what I see in astronomy – there is a dissonance between the astronomy culture of small groups, and the large-group physics collaboration culture. The classical style of astronomy is hardly dead and will not be for a long time. Making data available through archives ensures that astronomy will still be done in smaller groups. That is why the SDSS has more citations than the HST. The difference between astronomy and physics is that astronomy is still very data rich, and physics data poor.
Finally, I think we have the chance to invent new ways of giving credit. I dislike what is done in particle physics, and I got out of the field a LONG time ago as an undergrad at Stanford. When Brian and I organized the HZT in 1994, I made sure that we would work as a team but also assign credit where it is due. So we had two main rules (1) the entire team would work on the data and give the results to a different university group semester by semester (with the list drawn up at the start of the collaboration) and (2) the intellectual leader of that semester would be first author. That is why Brian, a postdoc and Adam who had just finished his PhD, were first authors on those papers. I don’t think that any such collaboration has ever had the *junior* people get as much credit for the discovery as in our HZT. The other team, the SCP, however was very much in the way particle physics is done.
This assignment of credit worked very well, until prizes granted the discovery to individuals and not the team. As expressed by Richard, this led to (as my thesis advisor would biblically say) “much gnashing of teeth.”
Last July Richard Panek came down to Chile to deliver 5 conferences on his book “The 4% Universe”. At lunch time at a restaurant in Santiago, Richard shared with us (José Maza, David Azócar, and I) his prediction about the Nobel Prize winners for the discovery of the accelerating Universe. Richard wrote on a napkin “50% for Saul Perlmutter, 25% for Brian Schmidt and 25% for Adam Riess”.
Not only Richard succeeded in guessing the winners of the Nobel Prize, but in his most recent LWON post he is absolutely right that the Nobel laureates represent the tip of the iceberg. In fact, Nick Suntzeff was crucial in founding with Brian Schmidt the High-Z team in 1994 (an idea that Nick had already proposed with the Calán/Tololo team in 1993 to the Cerro Tololo telescope allocation committee), Mark Phillips showed us in 1993 how to standardize the luminosities of Type Ia supernovae, and the Calán/Tololo (headed by José Maza, Nick Suntzeff, Mark Phillips, and myself) provided the calibration of the nearby Type Ia luminosities that Saul Perlmutter and Brian Schmidt ended up using in their discovery papers in 1998-1999 using distant supernovae.