Stephen Jay Gould stakes his claim to scientific greatness in his latest work. But, Mandy Garner writes, Gould's big idea may well have been superseded by other theories.

Once upon a time," says Henry Gee, senior editor at Nature magazine, "I commissioned a technical article by the palaeontologist, historian of science, evolutionary crusader, best-selling author and all-round good egg Stephen Jay Gould. For an article written by an academic, it was remarkable in that it was engagingly written; under the word limit; and arrived on the dot of the deadline. Closer inspection revealed a need for editing."

Gould is well known for his digression into personal and parenthetical comment, metaphor and anecdotes about anything from baseball to Gilbert and Sullivan. Gee says Gould's first paragraphs ran "true to form".

A tough editing hand was clearly needed, but accompanying the article was a covering letter in which Gould warned against cutting the prologue, "lest one find oneself, however inadvertently, walking over his grave, or words to that effect". Gee says: "Any young editor addressed in such terms by his literary hero would surely capitulate." He passed the text on to a more experienced editor who was "freer with the red pencil".

ÖйúAƬ

Gee doubts that he was the first editor to have been overawed by someone of Gould's stature. The Harvard University palaeontologist is famous for rarely allowing changes to his text, and critics would say that his tendency towards self-indulgence has grown at the same rate as his reputation. Gee reckons that Gould's bestselling 1989 book, Wonderful Life , was about a third longer than it need have been.

The Structure of Evolutionary Theory , Gould's latest and possibly last work, could be seen as the apotheosis of this tendency. The philosopher of science Michael Ruse views it as Gould's attempt to cement his status as "the major evolutionist of our time". At almost 1,500 pages, all written on a typewriter, it has been 20 years in the making.

ÖйúAƬ

Gould admits that he does not rewrite and that this book was not line-edited in the traditional way. Harvard University Press says some chapters were sent out, but not the whole book. Gould even admits that publishing such a lengthy tome could be seen as self-indulgent or arrogant. The opening section runs to 90 pages, almost a book in itself. Gould boasts that Structure , like Darwin's On the Origin of Species , is based on "one long argument", and it is to Darwin's book that it will inevitably be compared. Described as an attempt to reformulate Darwin's theory of evolution, it updates and revises the traditional canon for a new age.

The lack of editing speaks volumes about publishers' attitudes to the major players on the Darwinian stage. But, although lionised by many, Gould, a former president of the American Association for the Advancement of Science, has had a mixed reception among academics, particularly evolutionary biologists. Palaeontologists tend to regard him as one of the most influential thinkers of his time, even if some disagree with him, but evolutionary biologists are much more critical. Steven Pinker of the Massachusetts Institute of Technology, for example, has accused Gould of misunderstanding evolutionary psychology. Others are more personal, saying that his political beliefs and ideological agenda - his father was a leftwing activist, and Gould once famously remarked that he had learnt Marxism at "daddy's knee" - have blinded him.

Gould, who was born in September 1941, began to take on the core ideas of Darwinism during his undergraduate days studying geology and philosophy at Antioch College, Ohio, in the early 1960s. But it was in 1972 that he took the scientific world by storm with his paper on "punctuated equilibrium" written with Niles Eldredge.

Two years later, he started writing a monthly essay for Natural History magazine (the latest collection of some of these essays will be published this year). Gould's writing was well suited to the essay style - indeed many of the Natural History essays written over the past 30 years are peerlessly crafted. His first popular book, Ever Since Darwin , is a collection of essays. It was published in the same year, 1977, as his textbook Ontogeny and Phylogeny , which is widely considered his best work. Other books followed, including the 1982 book The Mismeasure of Man , which won the National Book Critics Circle Award in the US. It was in 1982 that he was diagnosed with abdominal mesothelioma. Despite the average survival time being just eight months, Gould, who has recently undergone treatment for a brain tumour, pulled through.

The experience, however, clearly left its mark. It was about this time that Gould started work on Structure , which brings together many of the ideas contained in previous works. There's the developmental biology of Ontogeny and Phylogeny , dusted off in the light of the modern fusion of genetics and embryology known as "evolutionary developmental biology" or "evo-devo". There's his critique of the "adaptationist" view that natural selection does not act on every feature of every living thing, but that the evolution of organisms is constrained by their history. And most of all, there's "punctuated equilibrium".

ÖйúAƬ

Punctuated equilibrium is a theory about a mode of evolution rather different from that usually associated with Darwinian evolution. Darwin supposed that species evolved from other species in a seamlessly gradual way. This view is challenged by the apparent lack of intermediate, or "transitional", forms in the fossil record. This problem concerned Darwin greatly, and he supposed that this deficiency might be attributed to the poor quality of the fossil record. After all, fossilisation is a chancy business. As only a vanishingly small proportion of all the organisms that have ever lived will be preserved as fossils, it is fair to say that many, if not most, transitional forms would not have been preserved or, to be more optimistic, awaited discovery.

In 1859, when the Origin was published, the geological heritage of many parts of the world was largely unknown, so a plain statement of ignorance was reasonable. More than a century later, however, much of the world's geology has been mapped, but the fossil record remains as free of transitional forms as ever.

Gould and Eldredge explained this by supposing that species vary in their rate of change. Most evolution happens very early in the history of a species - in the first few thousand years, perhaps, while the species sought clear blue water between itself and its progenitor. After this frenzied speciation, a species would settle down to a period of perhaps millions of years of relatively little change. And this is, seemingly, what one sees in the rocks. New species appear, apparently fully formed; persist for millions of years; and wink out. Speciation happens behind the scenes - very rapidly and in small populations - decreasing the chances of fossilisation to virtually nil. This model, Gould and Eldredge said, explains the evidence better than Darwin's "gradualistic" view.

ÖйúAƬ

From this, the original model of punctuated equilibrium, it became clear that the evolution of life, from the primeval blob to present biodiversity, cannot be explained wholly by the sum of the action of natural selection acting on individuals, extrapolated through geological time. Evolution, Gould and Eldredge claimed, is hierarchical, with natural selection acting at a variety of levels, from genes to entire species and groups of species. It is this hierarchical view of evolution, developed a quarter of a century ago and elaborated somewhat in Structure , that is the vessel of Gould's ambition.

But many would argue that Gould has been overtaken by events. After punctuated equilibrium, the next wave of theory to hit palaeontology, in the late 1970s, was cladistics - a method of determining evolutionary relationships entirely free from the problems of the patchiness of the fossil record. Because the fossil record is incomplete, one can never use it to trace lines of ancestry and descent in any way that is scientifically falsifiable, whether the descent is gradual or punctuated.

According to Gee, cladistics made punctuated equilibrium redundant a long time ago. He says: "Arguments from early 1970s mutton, no matter how they are dressed up as 21st-century lamb, are as outdated as the epicycles used to explain planetary movements in the pre-Copernican universe." Clearly Structure will continue to divide Gould's critics and will ensure the jury is still out on his long-term legacy.

Stephen Jay Gould's The Structure of Evolutionary Theory is published this week by Harvard University Press, ?.50.

ÖйúAƬ