2014年经济学人 天体物理 尘归尘土归土（在线收听）

Astrophysics

Dust to dust

A dramatic recent “discovery” in physics is looking rather dodgy

IN MARCH Chao-Lin Kuo, an astrophysicist at Stanford University, filmed himself knocking on the door of his colleague Andrei Linde. In the 1980s Dr Linde was one of several cosmologists who developed the theory of cosmic inflation, which holds that, in the first instants of its existence, the universe underwent a brief period of faster-than-light expansion.

Because inflation neatly cleaves several knotty problems in cosmology, many astrophysicists (though not all) subscribe to the theory. But direct, unambiguous evidence for it has been lacking. That was why Dr Kuo was visiting Dr Linde—to tell him that, thanks to the work of a telescope in Antarctica called BICEP-2, such evidence had now been found. After digesting the news, an emotional Dr Linde broke open a bottle of champagne to celebrate. The video has been viewed almost 3m times.

It now seems that Dr Kuo might have to make a new video, informing Dr Linde that he has wasted a bottle of bubbly. A paper just released by the team behind Planck, a European space telescope, casts serious doubt on the BICEP-2 result. What looked like a clear window back into the earliest moments of the universe might simply have been a faint glow from the diaphanous clouds of dust that exist between the stars.

The BICEP-2 team, led by John Kovac of Harvard University, had been studying the cosmic microwave background radiation (CMB)—a weak bath of radiation, left over from the Big Bang, that suffuses the universe. They were looking for evidence of primordial gravitational waves. These are ripples in the fabric of space, created, if the theory of inflation is correct, as the early universe was undergoing its post-creation growth spurt.

A dusty trail

Such waves should have left a distinctive, polarised mark imprinted upon the CMB. And, in a press conference on March 17th, that is exactly what the BICEP-2 team claimed to have found. It was the biggest news in physics since the discovery, in 2012, of the Higgs boson, and it was widely covered (including in The Economist). Not only would the BICEP result have confirmed the theory of inflation, but studying the gravitational waves it purported to have found would have given cosmologists a way to look back to the very earliest moments of the universe.

But gravitational waves are not the only things that BICEP-2 might have picked up. The Milky Way is filled with thin clouds of interstellar dust which, under the influence of the galaxy's magnetic field, scatter and polarise starlight. The BICEP team were confident that the contamination from the dust was small enough not to affect their detection of gravitational waves. But the behaviour of the dust is poorly understood, says Chris Lintott, an astronomer at Oxford University, and not everyone was convinced.

The Planck results suggest that they were right to be sceptical. The European telescope has just unveiled a map of dust density across the entire sky. It suggests that, contrary to the BICEP team's hopes, the signal from the dust is so strong that the telescope might well have seen no primordial gravity waves at all.

That, at least, is the most likely interpretation, but it is not the only one. The Planck team are careful to stop short of saying that their results are fatal to BICEP's claims, pointing out that applying their data to the BICEP results involves considerable “statistical and systematic uncertainties”. There are, in other words, a couple of glimmers of hope that signs of inflation have actually been seen. One is that the precise behaviour of the dust is still mysterious, which means the mathematical transformations used to apply Planck's data to BICEP's results may turn out to be incorrect. And even if those maths are sound, statistics may ride to the rescue—for if the amount of radiation from the dust is at the lowest end of Planck's estimates, a small gravitational-wave signal may survive.

Rowing back on a triumphant announcement about the first instants of creation may be a little embarrassing, but the saga is a useful reminder of how science works. There is no suggestion that anyone has behaved dishonourably. Admittedly, the BICEP team's original press conference looks, with hindsight, seriously overconfident. More information-sharing between the various gravitational wave-hunters, all of whom guard their data jealously, might have helped tone down the triumphalism. But science, ideally, proceeds by exactly this sort of good-faith argument and honourable squabbling—until the weight of evidence forces one side to admit defeat.

That could happen soon: the Planck and BICEP teams have pooled their data and are working on a joint paper, expected to be published in the next few months. Information from other gravity-wave hunting experiments—including some run by the BICEP team themselves—will shed extra light, too. It is not yet impossible that Dr Kovac and his colleagues will be proved right after all. But at this point it would take a brave cosmologist to bet on it.

From the print edition: Science and technology