I have just finished checking page proofs of an article that will appear in the 2015 Yearbook of Astronomy. The article is entitled “Ripples from the start of time?” and it discusses what had the potential to be one of the most important and exciting cosmological discoveries in decades: B-mode polarization of the cosmic microwave background. Earlier this year, the BICEP2 experiment claimed to have found just such polarization and argued that their measurements could only be explained in terms of primordial gravitational waves – ripples of space made large by inflation, an event that took place when the universe was only a trillionth of a trillionth of a trillionth of a second old.
The BICEP2 team made a bold claim, and bold claims require a lot of solid evidence before they can be accepted by the scientific community. Soon after the team made their announcement of B-mode polarization, scientists raised doubts about the interpretation of the measurements (though not of the measurements themselves: everyone acknowledges that the scientists involved here are extremely capable astronomers). One problem was that the BICEP2 experiment observed the sky at only one frequency: when you have only one data point, any curve can be made to go through it. When you observe the cosmic microwave background you need measurements at several different frequencies before you can be sure that your signal really does come from the distant cosmos and not somewhere nearby. A second, more pernicious, problem was that it was not at all clear that the BICEP2 team had properly accounted for dust.
The issue is that dust grains in the galaxy can polarize light – and in particular it can give rise to a B-mode pattern of polarization. If the BICEP2 team had underestimated the amount of dust emission then their interpretation of their observed signal had to be under suspicion. It’s why I added a question mark to the title of my article: BICEP2 might have been seeing a signal from the dawn of time, but it might not.
A recent paper by the Planck collaboration suggests that the BICEP2 result might well have been the result of dust. It turns out that there is much more dust in the area of sky observed by BICEP2 than was originally thought. That in turn means that the BICEP2 results are entirely consistent with observations of dust. This doesn’t mean that B-mode polarization of the cosmic microwave background does not exist, nor even that BICEP2 didn’t spot such polarization; but it does mean that we don’t need to invoke inflation in order to explain the BICEP2 results.
A joint paper by the Planck and BICEP2 teams, due for publication later this year, should clarify the situation further. But at the moment it seems that our dreams of being able to look back to the very start of the universe must be put on hold. Shame. Those dreams were beautiful while they lasted