Forget cosmology and quantum mechanics. A list of most important physics papers from the past century reveals condensed matter physics is where it's really at, says Philip Ball.
Lists. Don't we just love them? We are inundated by compilations of the greatest movies, the best pop singles, the most popular novels. A recent roll-call of Britain's 'top 100 public intellectuals' represents perhaps the most highbrow end of the market, and it may even be some comfort to scientists that biologist Richard Dawkins got the highest ranking.
But a list of the 100 physics papers that have had the highest impact over the past 110 years, announced by Sidney Redner of Boston University, Massachusetts, is more than an expression of personal opinion. The list forms part of a detailed analysis of citation statistics within the Physical Review journals since 1893. The study, which is preprinted on Arxiv1, provides a fascinating take on trends in modern physics and reveals the variable fates of scientific papers more generally.
As a window on twentieth-century physics, the study must be treated with caution: the US-based Physical Review may be the backbone of physics publishing, but the omission of papers and citations in other journals could severely compromise the significance of some of the findings. Still, appearing as it does just before the 'Year of Physics' in 2005, Redner's survey could hardly be better timed, and it yields some intriguing stories.
Centre of gravity
The most striking implication is evident from merely glancing at the list: physics' centre of gravity is not where the broader public (or even other scientists) thinks it is. There was, it seems, some eagerness to find a place in the aforementioned 'top intellectuals' list for Stephen Hawking, reflecting the widespread view that as a cosmologist he is at the heart of modern physics. But particle and fundamental physics in the Physical Review's top ten are notable only by their absence.
Neither is there much room (with one or two notable exceptions, to which I'll come) for that other facet of popular physics: fundamental quantum mechanics. The vast majority of papers in the top 100 are instead concerned with applied quantum theory: the quantum properties of condensed (particularly solid-state) matter.
It might seem less exciting than supernovas or theories of everything, but nothing, it seems, is more important to physicists than being able to calculate what energy states the electrons within materials have, and how they behave.
The top two papers2 on the list both address this problem. They are the foundational works in 'density functional theory', which provides an approximate way to figure out how an electron behaves in a fluid of other electrons. Both are co-authored by Walter Kohn, making him arguably the most influential physicist in the survey. Yet although he was awarded the 1998 Nobel Prize for his efforts in chemistry, because the method works for molecules too, he is hardly a household name.
The names throughout Redner's top 100 list tell a similar story: it is a roll-call of physicists' physicists. Yes, Albert Einstein and Richard Feynman are there, but so are Eugene Wigner, Phil Anderson, John Bardeen, Lars Onsager and Pierre-Gilles de Gennes. The list suggests not only that Kohn's Nobel Prize was absurdly overdue, but also that John C. Slater (another pioneer of quantum solid-state theory) was unjustly overlooked.
Alive or dead?
The analysis has other revealing tales to tell. The papers Redner picked are not simply the most cited. He argues that the impact of a paper also depends on its longevity, which is how long it continues to be cited for. So he defines the impact as the product of the number of citations and average age of those citations (the time between publication and citation).
Redner uses the citation histories to distinguish between two types of paper, which he brutally terms 'alive' and 'dead'. Papers die when they have few citations (fewer than 50 in Redner's classification) and when these appear shortly after the paper's publication. It is no surprise, perhaps, that most papers are destined for a quick death: of the 329,847 in the survey with at least one citation, nearly three-quarters have fewer than ten.
But among the highly cited papers, the histories are very diverse. Some stay hot for a remarkably long time. The number of citations for the two chart-topping papers co-authored by Kohn has grown steadily and is still increasing, even though they were published back in 1964 and 1965. The same is true for Phil Anderson's 1958 paper on localization in disordered systems, which remains a hot topic today.
Others maintain a steady or slowly growing citation rate: rather than 'hot', they are 'enduring'. The foremost example is Lars Onsager's 1944 paper on the two-dimensional Ising model, a lattice model of magnetism. The point here is that Onsager's system is very generic: all manner of phenomena can be modelled as particles interacting on a two-dimensional lattice, including ecosystems and (it now seems) many human social systems, such as electoral voting.
But the third type of long-lived paper is the most striking. These are publications that have a rather low rate of citation until suddenly, years later, their popularity explodes. All at once, they come into fashion.
Sometimes there is a technological driver behind this. A cluster of papers written in the 1950s on the magnetic properties of an obscure class of metal oxides could hardly have seemed at the time to be destined for 'classic' status. But then, in the late 1990s, it was found that these materials show an effect called colossal magnetoresistance, which was of great value for magnetic information storage, and references to the old work rocketed.
Ahead of his time
Einstein owes his only entry in the Physical Review list to one of these bursts of retrospective interest. He published many of his classic papers, such as those on relativity, in German, which is why they are among the most egregious omissions from the roll.
But in 1935 he collaborated with Boris Podolsky and Nathan Rosen at the Princeton Institute for Advanced Studies on apaper, published in Physical Review, in which he used a thought experiment to argue that quantum mechanics could not be a complete description of the world. The Einstein-Podolsky-Rosen (EPR) experiment4 purported to produce a logical absurdity when interpreted according to the standard view of quantum mechanics.
It was an idea way ahead of its time. The experiment was actually conducted only in the 1980s, and it revealed that Einstein was wrong: the quantum world really is that weird.
The tremendous upsurge in citations of the EPR paper has happened mostly over the past ten years or so, however. That is because the central notion that it invokes, quantum entanglement, is now key to the current interest in quantum information and quantum computing. Einstein's mental juggling has inadvertently provided the theoretical foundation of an emerging technology.
References
RednerS., Preprint"http://xxx.arxiv.org/abs/physics/0407137":http://xxx.arxiv.org/abs/physics/0407137 (2004).
KohnW. & ShamL. J. Phys. Rev. 140,A1133 - A1138(1965).
HohenbergP. & [author]KohnW.[/author]Phys. Rev. 136, B864(1964).
EinsteinA., PodolskyB. & [author]RosenN.[/author] Phys. Rev. 47, 777 -780(1935).
Related links
Rights and permissions
About this article
Cite this article
Ball, P. Unlikely chart-toppers. Nature (2004). https://doi.org/10.1038/news040726-16
Published:
DOI: https://doi.org/10.1038/news040726-16
