The egos have landed

Scientific papers are dispassionate and objective, right? Hugh Aldersey-Williams discovers that between the lines they are just as much about settling scores

Scientists call it "the literature", but the journal papers in which they announce their discoveries supposedly banish any literary artifice. Scientific writing is meant to be utterly objective - no waste, no frills, no personality. Compared with other forms of writing, it is, as one critic put it in the New York Times Book Review, "literary Novocaine".

It's easy to see why people might believe this. When undergraduate scientists learn to write up their experiments, they are told to organise their argument under headings of Method, Results and Discussion. To give expression to the objectivity that is central to the experimental procedure, authors are advised to use the passive voice: "such and such was done", not "we did such and such".

But the truth is rather different. On closer inspection, scientific papers turn out to contain seething subtexts that show scientists at their most human. They betray their authors' hopes and fears, their feelings of elation, or possibly of unease, at what they have found. They reveal envies and rivalries. They show signs of the manoeuvring needed to be the first to claim a discovery or to disguise a want of data for what is being claimed. Not bad for writing that is supposed to be entirely dispassionate.

Take what must be one of the most famous quotations from modern scientific literature. It comes from James Watson and Francis Crick's 1953 paper that first described the double-helix structure of DNA. "It has not escaped our notice," they write, "that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material." This humorous understatement, with its elaborate negative construction, breaks all the rules.

When you've just discovered the secret of life, you can perhaps be forgiven for getting carried away. But it turns out that all scientific papers have their moments. In Findings: Hidden Stories in First-Hand Accounts of Scientific Discovery, I look at a dozen or more major breakthroughs - or claimed breakthroughs - of the 20th century, from Max Planck's first hints of quantum theory in 1900 to the inconclusive evidence of fossil life on Mars produced by Nasa scientists in 1996.

Most of them shun passive objectivity and inject a clear authorial voice. These writers are desperate to ensure they are seen as the heroes of their own stories. And so, completely unabashed, Watson and Crick write "we have postulated..." at the critical point in their paper where they are drawing attention to the huge implications of the self-replicating double-helix molecule. Thomas Morgan, the American zoologist who in 1910 by means of an elegant series of experiments breeding generations of red- and white-eyed fruit flies discovered the link between sex and heredity, at first puts the flies at the centre of his story. It reads almost like a fairy tale, as a white-eyed "monster" appears in a population of red-eyed flies. Morgan waits until his conclusion to put himself at the centre with his first "I". In both cases, the strategy serves the authors well, strengthening the sense of connection the reader feels between discovery and discoverer.

The British physicist James Chadwick took just 700 words - half the length of this article - to announce his discovery of the neutron in 1932. Yet his paper is full of rhetorical flourishes glorifying British empirical science at the expense of his airy-fairy French rivals, Irène and Frédéric Joliot-Curie, who were fans of quantum theory. This reflected the long-held conviction at the Cavendish Laboratory at Cambridge University, where Chadwick worked under Ernest Rutherford, that neutrons were billiard ball-like particles. The French were equally convinced that they were more ethereal "hydrogen rays". Chadwick subtly mocks this vague conception throughout his paper, and even gets his rivals' names wrong. In his triumphant concluding sentence, he writes: "Up to the present, all the evidence is in favour of the neutron, while the quantum hypothesis can only be upheld if the conservation of energy and momentum be relinquished." By raising even the possibility of the loss of two of the great cornerstones of physics, Chadwick skillfully persuades us that reason is on his side.

But examining the texts of scientific papers can reveal serious flaws. In 1996, David McKay, a geologist at Nasa's Johnson Space Center, and a multi-disciplinary team of colleagues, seized the world's attention with the announcement that they had found evidence of fossil life on a Martian meteorite that had landed in Antarctica. Nasa organised a press conference at which not only the Nasa chief, Dan Goldin, but also President Clinton spoke. But the tone on this very public occasion was rather more confident than that in the scientists' own account of the story in their original paper, published in the American journal Science.

In marked contrast to the stylish writing of Morgan or Chadwick or Watson and Crick, the paper by McKay and his colleagues is wordy and woolly. It lacks the bravura that generally creeps into a really big announcement. Can it be that the authors' uneasy prose style is more revealing than what the words actually say? The claim to have found fossil life on Mars remains disputed.

The paper contains compelling specific evidence that all is not quite as it should be. McKay's team base their claim that Mars once supported life on three separate major observations of the meteorite, which they then connect to form their postulate. First, the meteorite contained carbonate globules - carbonates chemically identical to those in chalk or limestone formed by prehistoric life on earth. Second, there were hydrocarbon molecules similar to those found in fossil fuels on earth. And third, there were magnetic minerals just like those found associated with certain bacteria on earth.

The crux of the matter was the temperature at which the carbonates were formed. Other scientists had argued they had formed at high temperatures - too high to be consistent with life. McKay's team favoured a low-temperature formation, but had no evidence for it. They are driven to write: "If the globules are products of biologic[al] activity, a low-temperature formation would be indicated." This is their critical slip. The argument contains a false syllogism - in other words, it is circular. The true syllogism that would support the authors' desired outcome would run as follows: the globules form at low temperatures; low temperatures favour life; therefore the globules support the evidence for life. However, because they cannot state with confidence that the globules do form at low temperatures, the authors start with the major premise that "the globules are products of biologic activity"; only then can it be made to follow that the globules must have formed at low temperatures. Thus, the authors need the prior existence of life to create the impression of evidence to support their view that life existed.

This analysis arises solely from careful reading of the Science paper. With a bit of historical perspective, matters only get worse. For it emerges that McKay and his colleagues commit precisely the same error as a previous champion of life on Mars. A century before the Nasa announcement, the American amateur astronomer Percival Lowell looked at the "canals" on Mars recently spotted by the Italian Giovanni Schiaparelli, and pronounced that they were evidence of civilised life on the planet. He too used a circular argument to make his case. Beginning with the notion of civilisation on Mars, he argued that the Martians would require irrigation to bring water from the icy poles, and hence that this must explain the canals.

Why does this matter? Scientists take up positions when they argue for what they claim to have found. It is not simply a matter of reporting an observation and it being taken as scientific fact. Scientific truth takes time to settle, and new science is always contentious. To persuade others of their case, scientists must make their case as salesmen and politicians do - with rhetorical skill.

It's true that most of us do not read Nature as we might read the New York Times Book Review. But it is equally true that scientific papers are not completely opaque. They use the same language we all speak. Even if not all the detail is immediately understandable, we can appreciate something of how the scientists choose to argue their case. And if we can do that, we can begin to decide for ourselves the merits of that case.

'Hidden Stories in First-Hand Accounts of Scientific Discovery', £19.99, Lulox Books, www.luloxbooks.co.uk