Cosmic Confusion: It's How Science Gets Done
The reader may remember the news from around mid-March: a dramatic discovery made by scientists from the BICEP2 experiment, measuring what could be a signal from the very earliest times after the Big Bang, the closest we could hope to get to "creation" itself. It made the front page of The New York Times on March 17th: "Space Ripples Reveal Big Bang's Smoking Gun." Talk of a Nobel Prize and discovery of the century quickly ensued.
It may all have been too quick.
There's no doubt that scientists should hail such kinds of discovery with enthusiasm; sharing it with the public also makes sense, given that this type of research mostly comes from federal tax money. There is huge pressure on the team leaders to get the news out, both because of competition and because, well, because we humans are vain and love success. Who wouldn't want to have his name tagged along such a momentous discovery about the cosmos? After all, when you sign up for a career in the sciences your job is to get out there and explore. And explorers like to find things — the more relevant the better.
Three months after the breaking news, things have sobered up a bit.
Different scientists have cast doubt on the interpretation of the results. Not on the detection itself, mind you. The signal seems to be there, a sort of twist imprinted on the radiation left over when the first atoms formed, some 400,000 years after the Big Bang. The amazing thing is that theories predict that the twists originated even earlier, just a tiny fraction of a second after the Big Bang. The idea is that the universe underwent an ultra-fast and short-lived period of expansion right at the (perennially nebulous) beginning. This expansion caused the twisted ripples that affected the radiation much later.
BICEP2 claimed to have measured the twists, providing evidence for the ultra-fast expansion and the ripples, which happen to be ripples on the geometry of space itself. (Here are a couple of earlier posts where I explain the physics in more detail: "Listening To The Echoes Of Creation" and "The Truth Is Out There In The Fog.")
All we can do when we attempt to understand nature is to measure things up the best way we can. We catch all sorts of different signals with our tools of exploration and then construct models to make sense of them. What makes it hard is that nature doesn't tell us what's causing the signal. It could be caused from the source are targeting. Or it could be from a completely different source, from this or that phenomenon. To discriminate between potential sources, that is, to zero in on the correct source of the signal, is not easy.
This is what's going on right now with the signal from BICEP2. Is it primordial, something that takes us back to a very young universe, or is it coming from a more mundane source, like polarized radiation from dust grains floating about in the interstellar medium?
I was in a conference last week at the University of Chicago where one of the BICEP2 team leaders, Chao-Lin Kuo from Stanford, explained that it is indeed possible that the signal they measured did not come from the early universe. The consensus is that they have to wait and see. Wait and see means data will be coming in from different experiments. This is very important. The scientific community will only embrace a discovery of this magnitude after corroboration from one or more separate and independent groups.
Interestingly, the official research paper relating the discovery just came out last week in the prestigious journal Physical Review Letters. Already in the abstract the authors show caution. After discussing the possible sources of signal masking, they state:
"However, these models are not sufficiently constrained by external public data to exclude the possibility of dust emission bright enough to explain the entire excess signal."
The entire excess signal, that is, the discovery. In a "note added" before publication, the authors mention recent results from the Planck satellite collaboration that map the contribution from polarized dust in a huge part of the sky — but not the one targeted by BICEP2. However, and this is where the tension lies, the signals measured from Planck, exclusively coming from polarized dust, could reproduce the BICEP2 discovery. So, we need more data in the region of the sky that BICEP2 covered to check things out.
What do we learn from this? First, that the results from BICEP2 were amazing, regardless of their interpretation. It was a tremendous technological feat to register the twisted radiation signals scattered in a patch of the sky from detectors at the South Pole. Second, that team leaders should be a bit more cautious before going public. The BICEP2 team was very careful in their data analysis, in searching for possible sources that could mimic the polarization from gravitational waves. But more restraint might have been good, including from the theorists that lit the fireworks and opened the champagne way too quickly. Third, that modern science works communally, that is, through the vetting of colleagues. It is an amazingly effective model of competition-driven international collaboration, providing striking proof of what humans are capable off when they work together.
Confusion is an essential part of the creative process. Science is about figuring things out; that usually doesn't include finding the answer wrapped up and ready to go. It's rarely that easy. Amazing claims require amazing evidence. And patience, lots of it.