The evidence for gravitational waves revealing cosmic inflation from the early moments of the universe is deflating. It seems increasingly likely that what looked like a triumph of scientific discovery is instead a triumph of the scientific method, an example of how scientists find and correct each others' mistakes.
In what appeared to be the scientific announcement of the year, astrophysicists using the Background Imaging of Cosmic Extragalactic Polarization (BICEP) telescope at the South Pole claimed to have found patterns in the polarization of radiation from the early universe. In Physical Review Letters, they argued the patterns were indicative of a rapid acceleration in the expansion of the universe at the time.
This was in keeping with the Inflation model of the early universe. Indeed, the strength of the signal they found was right at the highest end of what theorists had postulated for the amount of inflation that might have occurred. Excitement was intense. If correct, the finding confirmed not one but two of the most important theories in physics: The inflationary big bang and general relativity.
Moreover, such a find means being able to probe a period of the universe's formation hidden from existing telescopes.
However, doubts emerged fairly quickly concerning the possibility that the BICEP team had misunderstood data from the European Space Agency's Planck space telescope team on microwave emissions from dust in other galaxies. Since these emissions needed to be removed to get a picture of the Cosmic Microwave Background BICEP were trying to study, knowing what needed to be subtracted was crucial to getting the right result. BICEP used a part of the sky where interference was thought to be small, but perhaps not small enough.
Science enthusiasts were told we would need to wait until October for the Planck team to come out with their own findings on the state of polarization of the cosmic background, but in June the team behind the original announcement admitted, “What we pointed out in our paper is that pattern is just as consistent with the galactic dust radiations as with gravitational waves.”
The Planck team's paper is not yet out, but they've already dealt a major blow to BICEP, and inflation. A paper to be published in Astronomy and Astrophysics and available in preprint at Arxiv finds a lot more emissions from dust away from the plane of the galaxy than previously thought. The authors write, “We show that even in the faintest dust-emitting regions there are no “clean” windows in the sky where primordial CMB B-mode polarization measurements could be made without subtraction of foreground emission.”
Looking at the area studied by BICEP, the Plank team concluded uncertainty in polarization measurements is of similar size to the signal thought to come from the early universe. They state that these uncertainties can be reduced with further analysis, but a margin of error that large leaves the original claim looking weak.