Jun 20 2014

Inflation Evidence Questioned

Published by under Astronomy
Comments: 4

A critical part of skeptical outreach is teaching the public about how science works. Surveys of scientific literacy generally have dismal outcomes, but they also generally focus on knowledge about the findings of science, and not so much on the process of science. My personal experience from engaging with the public in multiple venues over decades is that those who are critical or suspicious of science generally are laboring under a gross misunderstanding of how science operates.

Actually it’s not quite accurate to talk about “science,” and that is not how I think about or evaluate scientific claims. Rather, the global scientific community has a certain culture and norms of acceptable behavior. Each country, however, has their own subculture and may have problems or failings specific to them. China, for example, apparently can only publish positive studies about acupuncture, betraying a national bias that calls into question any acupuncture study originating from that country.

Each scientific discipline also has its own subculture. Some professions and specialties are more rigorous than others. Further, each institution, lab, and researcher has their own culture and behavior.

When I evaluate a claim, therefore, I try to put it as much as possible into context – is the discipline rigorous, is the research up to international standards, do the particular researchers/institution/publishing journal have a good track record? It is just as much folly to accept all science as rigorous as it is to condemn all science as fraudulent or biased.

It is good to note, however, that standards for scientific rigor are generally high within the respected institutions and disciplines of science. Also, the larger the scientific community, the more likely it is that local researcher or even institutional quirks and biases will average out. This is why it is important for findings to be published in peer-reviewed journals and subsequently picked over by skeptical colleagues. Not until a study has gone through this meat-grinder do I even pay much attention to the results.

The latest example of this is an excellent lesson in how science is supposed to, and often does, work. In March of this year physicists announced that they had discovered evidence of the cosmic inflationary model. This model hypothesizes that in the first trillionth of a trillionth or a trillionth of a second after the Big Bang, the universe underwent a burst of hyperinflation.

One primary reason for this hypothesis is to explain why the universe looks so uniform. In all directions, the cosmic microwave background  is very smooth, rather than wrinkled. The idea is that the rapid inflation would have smoothed out any early wrinkles in the universe (the metaphor often used to describe this is flattening out a crumpled piece of paper).

A nice idea in science is not that useful, however, unless it makes specific testable predictions. The inflationary hypothesis predicted that there would be a specific pattern of polarization in the cosmic microwave background (CMB), so-called B-mode polarization.

That was the announcement this past March – astronomers had found the predicted B-mode polarization in the CMB, apparently confirming the inflationary hypothesis. This was hailed as one of the greatest discoveries in cosmology perhaps since the discovery of the accelerating expansion of the universe in the 1990s.

Before the scientists could be done patting themselves on the back, however, there were already murmurs of healthy skepticism. Are the characteristic swirls of B-mode polarization real, are they an artifact, are they due to galactic dust rather than the true CMB?

Now that the paper has been published in a peer-reviewed journal, the scientific vultures (and I mean that with respect) are lining up to pick the meat off the bones of this evidence. The BBC reports:

At his lecture at University College London, Prof Pryke explained his team’s lowered confidence: “Real data from Planck are indicating that our dust models are underestimates. So the prior knowledge on the level of dust at these latitudes, in our field, has gone up; and so the confidence that there is a gravitational wave component has gone down. Quantifying that is a very hard thing to do. But data trumps models.”

Data trumps models. That is the way science is supposed to work. The noise in the CMB generated from galactic dust has to be removed to see if the B-mode polarization is left behind, which would be evidence of gravitational waves originating from cosmic inflation. The primary criticism of the data is that the researchers did not adequately remove the dust artifact, and if they do then perhaps the signal will go away.

For now the researchers are standing behind their results, but admit that their confidence has diminished. What is likely to happen now is that we will have a few years of controversy until other teams are able to independently replicate the research, and either confirm or refute the earlier findings.

Either way, science will have advanced. Negative findings are still findings, and the help us move forward to figuring out the history and nature of the universe.

This is a great reminder, however, that even a high profile and exciting finding like this one still has to go through the process of peer-review, and still may be destroyed in the end by a careful and skeptical approach to the evidence.

This is an excellent counterexample to those critics who would dismiss any scientific findings they don’t like by impugning the very process of science itself.

4 responses so far