More than the sum of its particles

At the end of the 19th century it seemed that physics was nearly complete except for a few loose ends, yet, starting in 1895 with Rontgen's discovery of X-rays, a sea change occurred. By 1906, Lorentz and Einstein had formulated special relativity, Planck had initiated quantum theory, Thomson had discovered the electron and Becquerel radioactivity, de Forest had invented the triode and Marconi had transmitted radio signals across the Atlantic. In experimental and theoretical physics and related technology, the era of modern physics had begun; and so 1900 is an excellent place to start a new chapter in the history of physics.

Most of this review deals briefly with the contents of the 26 sections where related topics have been grouped together. At the end, we look back over the century. Thirty-six authors covering such a vast subject, in more than 2,000 pages and three solid volumes, present a solitary reviewer with an impossible task, so I am grateful to Mike Glazer, Raymond Hide, Tom Mullin and Don Perkins for dealing with those chapters where I felt that my knowledge was inadequate.

The three volumes are easy to handle, there are excellent subject and name indices and extensive lists of references at the ends of most of the chapters. Who, then, is it all intended for? Probably no one will want to read every article, though anyone pretending to be a physicist should be able to do so. Both I and my colleagues feel that some chapters should be read by everyone embarking on a career in physics. Thus, even if not many will buy it for their personal use, Twentieth Century Physics ought to be in every library used by graduate or undergraduate students. If it falls between two stools - neither a reference book nor exactly science history - this is not a bad thing. Rather we should treat it as providing a sense of perspective and an inkling of how physics, at the end of the century, reached its present state. The preface admits one very serious weakness: there is nothing about electronics. Surprisingly the editors could not find anyone to take it on, and so the major experimental technique and the archetypal technology of the 20th century is missing.

After reading Brian Pippard's clear account of the situation in 1900 my hopes were high, but they fell steadily during the next three chapters: "Atoms and nuclei" by Abraham Pais, "Quanta" by Helmut Rechenberg and "Relativity" by John Stachel. Both the general reader and the student of physics would want to know what all the furore in the first quarter of the century was about, yet there seemed too many physicists' names, too much time spent allocating credit and too little physics. Some idea of why I found these chapters so unsatisfying can be gleaned from three examples. In chapter two we find: "Once upon a time there lived a man in the town of Heidelberg named Gustav Robert Kirchhoff (1824-87). He became one of the very few 19th-century physicists to make fundamental contributions both to experiment and theory" - and this in the century of Maxwell, Helmholtz, Kelvin and Rayleigh! Then in chapter three we come across ". . . the properties of . . . semiconductors: their bands do not overlap at zero temperature but do so increasingly for higher temperatures". In chapter four, after 35 pages on special relativity, quite unnecessarily concerned with assigning priority to Einstein or Lorentz (whose transformation is not presented), there is a page and a quarter about its experimental verification, which contains nothing about time dilation and the muon lifetime, nor how, in Dirac's hands, it led to an explanation of "all of chemistry and most of physics". One could not recommend these three chapters to physics students or the general reader.

My expectations revived with "Nuclear forces" by Laurie M. Brown, to be discussed later, and clearly Mike Glazer felt the same about the next two chapters. In "Solid state structure analysis" by William Cochran, after an account of the impact on science of the discovery of X-ray diffraction by crystals, and some interesting potted biographies of the pioneers such as von Laue and Lawrence Bragg, he found all the main techniques clearly described, together with their applications to proteins, viruses, phase transitions, h.t. superconductors and so on. Glazer's only reservation was that the text was rather short on modern techniques such as synchrotron radiation. "Vibrations and spin waves in crystals" by R. A. Cowley and Brian Pippard extends structure analysis, principally through the use of neutron diffraction, to atomic vibrations, with all their consequences for thermal and transport properties. The long stretch from the elementary ideas about specific heats of Einstein and Debye, to spin waves and phase transitions, is treated without mathematical obfuscation, and there is a sketch of the experimental techniques, from the calorimetry of Eucken, Nernst, Lindemann and Simon, to Brockhouse's pioneering neutron work at Chalk River.

Pippard begins "Electrons in solids" with sensible remarks about the limited quantitative predictive powers of theory, and the dependence of progress on the availability of pure and nearly perfect crystals. This rewarding article, by someone who has made many important contributions to the subject, would be a useful introduction for graduate students.

"Superfluids and superconductors" by A. J. Leggett is a first-rate account of a difficult subject. Though none the wiser, I felt much better informed when I had finished it, but it would have been more useful with more references.

K. W. H. Stevens has made important contributions to the theory of magnetism, but in "Magnetism" he manages to present the theory of almost every kind of magnetic effect using hardly a single equation. This is followed by a substantial and helpful section on the applications of magnetism.

"Physics of materials" by R. W. Cahn is nearly all about metallurgy. Other aspects of material science get short shrift, with nothing about adhesion and adhesives and very little about optical, magnetic or electrical materials.

"Equilibrium thermodynamics and statistical mechanics" by Cyril Domb, reviewed by Glazer, deals with the developments in this pre-eminently 19th-century subject due to quantum mechanics. It then describes their connection with the Third Law and low-temperature physics. It also covers Boltzmann's ideas, rather neglected at the beginning of the century. There is a substantial discussion (including both Landau's macroscopic treatment and the microscopic treatments associated with Onsager, Kadanoff and Wilson) of phase transitions, a subject close to Domb's heart and of primary concern throughout our century. "Non-equilibrium statistical mechanics" by Max Dresden introduces us to the controversies surrounding this subject. Dresden has a light touch and this chapter is a good read and ought to be of general interest to physicists from other fields.

It is exhausting to be led through 80 years of "Atomic and molecular physics" in 83 pages, even by Ugo Fano. He uses the very minimum of equations, and provides a splendid guided tour of the subject, from the days of Bohr to the discoveries made since lasers revolutionised spectroscopy.

"Optical and optoelectronic physics" by R. G. W. Brown and E. R. Pike is a spectacularly comprehensive and thorough survey of one of the most continuously active 20th-century fields of experimental and theoretical research, and its host of technical applications. Some idea of the breadth of the treatment can be gauged from the 873 references at the end of the chapter. For years it should be a point of departure for anyone starting work in optics. It also gives a lively sense of the progress from Drude's text, at the beginning of the century, to the present day.

"Nuclear forces" by Laurie M. Brown is a relatively short but delightful article. It reminds us of the important role played by cosmic ray studies in the years before high-energy accelerators became available, and ends with C. F. Powell's development of photographic-emulsion particle-detectors, and the experiments in 1947 that led to the discovery of the pion, initiating what we now call elementary particle physics. Another short and readable article, "Nuclear dynamics", by David M. Brink, deals with a more clearly defined topic: how protons and neutrons behave in nuclei, how protons change into neutrons and vice versa and how they get rid of excess energy. For me it cleared up many odd problems left over from days and nights spent observing gamma rays emitted by oriented nuclei.

"Elementary particle physics since 1950", by Val L. Fitch and Jonathan L. Rosner and reviewed by Don Perkins, is a much longer article, as befits a topic that many physicists believe is the frontier of 20th-century physics. It traces its development over the past 60 years or so. The historical sequence of events and the atmosphere of the early days are well described, and the comprehensive list of more than 500 papers and conference reports is an outstanding feature. Beginning with quantum electrodynamics and its experimental verification (Lamb shift and so on), it moves on to how cosmic ray studies in the 1930s and 1940s stimulated the building of bigger and bigger accelerators. There is a good discussion of the design of these machines and the associated detectors, though rather a shortage of explanatory diagrams. Generally the article provides a thorough nontechnical account of the development of the quark concept, the renormalisable field theories of interactions between quarks and leptons, and the "Standard model". There are a few errors, an occasional lapse into total obscurity (for example in discussing the dependence of the neutral pion decay rate on the number of quark colours) and curiously little space is devoted to the mass of work on the Zo boson, showing that there are exactly three neutrino flavours to match the three pairs of quark flavours. The few pages devoted to cosmological topics convey little sense that accelerator experiments probe conditions in the very early stages of the universe. Despite these shortcomings, this is an extremely well-balanced and readable account and will surely remain a standard work of reference for years to come.

In "Astrophysics and cosmology", Malcolm S. Longair has produced a remarkably thorough and gripping account of progress in our knowledge of the universe, and the experimental techniques that have made it possible. But, with its 440 references, it may not make enough concessions to attract the general reader. It certainly avoids the popular cosmological science fiction that betrays the integrity of physics, and it should hold amateur astronomers and physics students spellbound.

"Units, standards and constants" by Arlie Bailey, with 126 references, is a useful account of the growth of international and national standards institutions, and of the progression from the ad hoc definitions at the beginning of the century to the current definitions of almost every unit, except the kilogram, in terms of atomic phenomena such as the Josephson effect and the quantum Hall effect.

We now come to a group of subjects in which the Planck constant makes only a fleeting appearance. Not surprisingly Richard F. Post, in a 70-page article, "Plasma physics" gives the first half of the century only ten pages at the beginning, and the upper atmosphere only six pages at the end. Plasma physics today, whatever its geophysical and cosmological significance, cannot escape being dominated by the worldwide attempts to generate energy by controlling nuclear fusion. This, to quote Post, "has had to live between the rock of Bohm diffusion and the hard place of impatience on the part of the body politic with the fact that no practical fusion reactor has resulted from over 40 years of internationally supported research". Despite this, he has produced an informative account of both progress and disappointment in this important field. There must be a lesson here for those who think that "mission-oriented" research will always deliver the goods.

The next three papers were reviewed by Tom Mullin, beginning with "Fluid mechanics" by James Lighthill, a clear authoritative account by one of the giants of the subject of its early development and the profound effect of Prandtl's discovery of the boundary layer. This is linked to later applications such as supersonic flight and the use of computational fluid dynamics in weather forecasting. Having said that, it is a slightly disappointing article, not containing enough about the nature of fluid flows. The emphasis on the use of modelling techniques in practical problems gives the impression that the physics is now all understood, so that computation will inevitably reveal all. There is, for example, little reference to modern developments in dynamic systems theory that suggest that predictability, in computing nonlinear equations, is far from certain. As if to compound the issue, the article continually mixes the concepts of chaos with the outstanding problem of classical physics, turbulence. Most modern work draws a clear line between the limited temporal nature of the former and the broad scales in both the time and spatial signatures of the latter.

"Computer-generated physics" by M. J. Feigenbaum is a fascinating article by one of the founders of the subject, a personal account of the development of the computer age and the uses and abuses of computers in physics. A riveting account of the detection of universality in period doubling describes the trials and tribulations of Feigenbaum's journey to this discovery.

"Soft matter" by P. G. deGennes describes the use of incisive physical thought to summarise the common features of a complicated range of materials, ranging from the vulcanisation of rubber to screening effects in colloids. This is a branch of physics where weak perturbations to a molecular system can cause long-range order. A familiar example is the effect of a small electric field on the liquid crystals used in displays. This brief chapter whets the appetite and should encourage the reader to use the list of references.

"Electron beam instruments" by T. Mulvey discusses electron microscopes but contains nothing about high-current beams, klystrons and travelling wave tubes or about the camera tubes and cathode ray tubes that made television possible.

"Medical physics" by J. R. Mallard begins like "modern" physics with Rontgen's discovery of X-rays, whose use was almost synonymous with medical physics in the first half of the century. This is followed by electrocardiography, medical electronics and so on, all the way to the growing uses of magnetic resonance imaging. Curiously, there is almost nothing about lasers or fibre optics, but the overwhelming impression is still of the importance of physical instruments and techniques in diagnosis and treatment.

Raymond Hide reviewed "Geophysics" by S. G. Brush and C. S. Gillmor who have taken a broad, though not encyclopedic view of their subject. They discuss in some detail the origin and age of the earth, the discovery of its liquid metallic core and the origin of the geomagnetic field, continental drift and plate tectonics, the upper atmosphere and the magnetosphere. By sticking to the subjects they know best they are able to give fascinating, blow-by-blow details, but several more authors and chapters would be needed to do justice to the role played by geophysics in 20th-century physics and how it kept what Weisskopf called the "inner frontiers" of physics alive. Indeed a meteorologist, Lorenz, caused a profound change of outlook in late 20th-century physics, by discovering deterministic chaos.

The three essays "Reflections on 20th-century physics" that end the work are each very different. Philip Anderson sees the most hope for the future in our growing capacity to deal with more complex systems, but Steven Weinberg seeks an ever more fundamental theory at the reductionist frontier of physics. John Ziman, worries about the relation of physicists to the body politic, so that he ends with " I and exercising their political and managerial skills to steer the enterprise safely as it forges ahead through very dangerous seas". All three, however, are united in thinking that we are seeing the passing of an era as well as a century, but they do not seem to share the view of their 19th-century predecessors that there is nothing left to discover. In all three there is also an element of nostalgia, recalling Pippard's perceptive remark in the first chapter that "physicists who entered the profession after 1960 have had no firsthand experience of a time when the number of papers relevant to their special interest was not too large to follow closely, and when the leading figures all seemed to know each other personally".

The three essayists, being theorists, seem to ignore another possible future, that experimental physicists will continue to devise new ways of investigating the physical world, that these new techniques will be adopted in other sciences and that it will be these applications that become the frontiers of science, not physics itself. We have already seen how X-ray crystallography has changed the nature of biology, and how magnetic resonance is changing our understanding of the brain. We should remember Maxwell's warning and not "assume that the physical science of the future is a mere magnified image of that of the past". The authors are all physicists at the University of Oxford.