Still confident of being surprised

It can be no accident, in an academic world that restricts and often rejects scientists who stray from the rectilinear rigours of their sub-sub-sub-fields, that John Maddox took early leave of the university system to become (not once but twice) editor-in-chief of Nature , the world's most highly respected scientific journal. When Maddox left the physics department of the University of Manchester in 1955 to write for The Manchester Guardian he was already on a trajectory to become, in the field of science, arguably the most widely educated person in the world.

The breadth of his personal interests and the depth of his scholarly inquiry render What Remains to be Discovered dazzling in its coverage of the major facets of funded science: the origin of the universe, the origin of life, the workings of the mind, the use of mathematics in scientific description, the alteration of climate by humans, and the problems of scientific inquiry itself.

While pointing out impressive scientific successes, Maddox derides the hubris of claims that "science has ended". For anyone who follows even a minuscule part of the great search for truth based on evidence, the scientific quest with all its foibles is just beginning. Quantum theory's success in predicting how charged particles in a vacuum interact with photons has surprised even its ardent practitioners. Experiment agrees with theory to within a few parts per 100 million. In an evacuated chamber containing two charged flat plates, the supposedly empty space between the plates exerts a pressure (the so-called Casimir effect). This and many other observations are accounted for by quantum electrodynamics (QED).

In 1996, 18 "galactic clumps" (blue galactic objects all smaller than our own and other close galaxies) were found in a small patch of sky by University of Arizona astronomers using the Hubble telescope. Such observations, of billions of early stars prior to galaxy formation, take us two thirds of the way back to the big bang (if there was a big bang) and remind us that science is far from enjoying the complete universe's narrative or even a "natural history of galaxy formation".

Does the existence of black holes help account for the general observation in galaxies that stars are "travelling faster than can be accounted for by the mass of visible stars?" Here Maddox speculates that the "idea that the universe began in a single event, the big bang, will be found false". We have been viewing the universe by chinks of light through a blindfold, he says, but that will all change. In half a century, he surmises, cosmologists will reject the "once and for all universe of Genesis, or [Alan H.] Guth's equivalent, as improbable". Indeed, he appropriately warns us in his discussion of "theories of everything" that "it will be time enough to talk about theories of everything when we know what everything is".

Beyond physics and astronomy, in parts two and three of the book, lies life: its origin as replication molecules on a prebiotic earth, and its organisation as cells about which molecular biology has created a "huge amount of data, all potentially interlinked", where intricacy abounds. But what, asks Maddox, "is to be done to make the data intelligible?" From these general questions, about which Maddox provides a wealth of detailed illumination, he proceeds to his next set of preoccupations. "Understanding how the brain works", he admits, is unfortunately "a more distant goal". The third and last section of the book covers the thinking machines, computers. In a chapter called "The numbers game" he waxes on the glories and limitations of mathematical description. For Maddox, mathematics is our greatest invention.

The last portion of the book, called "Avoidance of calamity", deals with epidemics and global climate change. Here appears some Maddox musing on how communities of scientific researchers publish, communicate, please their funding agencies and, pressured in other ways, interact socially. "Research communities are everywhere under pressure to become 'relevant'," Maddox asserts, which is "usually understood to mean that they should assist with national competitiveness in the production and sale of tangible traded goods." Although the frontiers of science described in this book are limited to the "pure" (=basic, =academic) science, Maddox lauds applied science for dramatically changing and improving the lives of people.

Inspection of the recent history of widely held scientific ideas reveals back alleys repaved, dead ends, the implosion of expensive, grandiose, unfounded claims, and simple errors. Well aware of the limitations of science, Maddox warns us that "a brief listing of the foci of our present ignorance is incomplete". Science is full of surprises: another century from now, "people will be occupied with questions we do not yet have the wit to ask".

The success of science engenders "a corrosive impatience". The book ends on an optimistic personal statement that all of us can learn directly from evidence. Maddox predicts that much more remains to be learned than has already been discovered. He hopes that open, active, cooperative scientific investigation, along with mathematical reconstruction and strong inference, will occupy our lively minds and those of our children and their children, "perhaps even to the rest of time".

Inevitably, the broad brush has a tendency to miss the minutiae by which scientific knowledge is compelled to expand. And when Maddox describes the Gaia hypothesis as "wish fulfilment of a kind" and uses the tired zoological liturgies of "adaptation" and "natural selection" as if they explained the origins of species and other evolutionary novelty, his reliance on popular peers rather than primary literature is apparent. His most serious flaw in these areas is the same as the flaw rampant in the community of scientific scholars he so assiduously and accurately reviews. Here he lacks a true historical perspective, especially the "deep time" geological one that relates natural selection to environmental chemistry, in the manner of Vernadsky and Lovelock. Yet by the communication of his immense knowledge about what we think we know and exactly when and how we learned it, Maddox greatly aids both scientists and historians of 20th-century science, as he reveals so many different current approaches to the vastness of what remains to be discovered.

Lynn Margulis is professor in the department of geosciences, University of Massachusetts, Amherst, United States.