The recent brouhaha over whether there's compelling evidence for life on Mars offers a stark lesson about research life: a major scientific discovery is a temptress as beguiling, and as dangerous, as the Sirens that beckoned Ulysses.
To learn something both important and new not only guarantees a scientist steady employment and a wall-load of awards; it can permanently fix his name in the big book of human achievement. It's immortality, of a sort.
On the other hand, a claim that turns out to be mistaken is often an indelible black mark, leading to criticism, ridicule, and a brisk ride to oblivion.
Many people outside the scientific establishment have a Hollywood view of research. For them, discovery consists of a "eureka" moment when a scientist - impelled by brilliant insight or too much French roast - will suddenly cover blackboards with fresh mathematica or throw a knife switch in the lab, and moments later step back to confront some new, profound truth.
In fact, much discovery is tentative. The data suggest something, but they are seldom thoroughly unambiguous except, perhaps, in retrospect.
So a researcher has to make a choice of whether to "interpret" the data and gamble on discovery, or keep quiet, continue the slog, and risk being trumped by a competitor. In other words, there's a fine balance between discreetly jumping the gun and being out of the race.
The right choice is seldom dead obvious. In 1989, chemists Stanley Pons and Martin Fleishman claimed they had provoked cold fusion in their lab. What they actually provoked was cold rebuke from their peers. In contrast, Edwin Hubble pored over his plot of galaxy spectra in 1929, and decided that the universe was expanding linearly. If you look at those data yourself, you'll appreciate the degree of inspiration (or imagination) required to dare that conclusion. More than a decade before Hubble, astronomer Vesto Slipher had already measured the recession velocities of a fistful of galaxies, and found that most of them were receding. If he had been a bit more audacious, NASA might be debating the merits of a Slipher Service Mission to save our premier space telescope.
When the first planets were found around other suns a decade ago, there were serious people in the astronomical community who believed that the claimants had merely observed double stars, not planetary systems. The discoverers held fast, however, and were eventually guaranteed ink in every Astronomy 101 textbook from now until the Sun goes nova. They did this despite a sobering precedent: in the 1960s and 1970s astronomer Peter van de Kamp had gone public with the news that two nearby stellar systems, including Barnard's star, danced back and forth on the sky, proving they hosted planets. Unfortunately for van de Kamp, the dancing was caused by periodic maintenance of his telescope, not by distant worlds. Sic transit gloria mundi.
There's also the problem of interpreting data based on preconceived notions. Facts are, research often proceeds in an attempt to prove what you think is true. In polite parlance, this is admired as "proceeding on a hunch." Of course, when your hunch is hogwash, the results are rubbish. Columbus departed this world still believing his ships had landed in the Indies, because that's where he thought he'd land. Had his notions been more open, we'd be calling it the United States of Columbia.
It's hard to imagine a discovery more grand than finding intelligent life in space. However, SETI scientists have assured both themselves and anyone who asks that they will meticulously check any candidate signals picked up by their telescopes, making sure they are truly extraterrestrial and truly technological. They take to heart the lesson of the pulsars: signals that, at first, seemed to be the handiwork of Little Green Men - but turned out to be a previously undiscovered creature in Nature's celestial bestiary. They remember CTA-102, a variable, cosmic radio source that Russian researchers thought might be the noisy broadcast of an advanced society. CTA-102 was a quasar.
Nonetheless, good intentions are easy, and deciding on the data is hard. So when SETI researchers find an interesting signal, they don't immediately flush with adrenaline and shout from the shingles. They know that any signal will need extensive confirmation to rule out such mundane origins as a telecommunication satellite, local interference, or squirrelly software.
But at some point there will be a need to decide which way this teeter-totter is coming down. A wrong call could simply make you grist for Jay Leno's monologue. But a delay might mean that your ticket to Stockholm will be given to someone else.
It will, no doubt, be a tough call.