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

4 thoughts on “Inflation Evidence Questioned”

  1. nybgrus says:

    I actually made a comment over at reddit on this exact topic. Specifically it was in regards to this article in Nature by Paul Steinhardt. While the idea of picking apart and trying to demonstrate problems with the BICEP2 data is precisely the way science should work, I found Steinhardt’s article to be more of a screed than a legitimate scientific criticism. The topic, while something I follow and know a fair bit about, is far outside my expertise. So I did what any good skeptic should – I looked to trusted experts and their analysis on the topic to see what they had to say. I looked at multiple sources and even tried to look at a primary source paper but it was way over my head. So I was left only being able to say what I wrote in that Reddit comment (which I would reproduce here to save people the click, but my hyperlinking would be lost and that is where I reference my sources; so if anyone is interested to read more on the topic the links are there).

  2. Jerry in Colorado says:

    Excellent blog Steven. I will save it for my friends who don’t understand the validity( or lack of) of scientific findings to show how science works.

  3. Physics has generally been an early adopter of more rigor, e.g. arXiv as a way to solicit feedback quickly. But the creation of facilities like PubPeer and PubMed Commons means that post-publication peer review is growing in many other (all?) disciplines, too. As a practitioner I think this is great, possibly the best new tool for sharing science since e-mail. We are going to see ever quicker responses to papers, pre-publication peer reviewed or not. Done correctly, it should save a huge amount of time. (Although as with e-mail, it’s all in the doing!) Hundreds of extra eyeballs and a record of their efforts will be invaluable. Those looking to cast aspersions will surely have more grist for their mills, as we make our sausages ever more publicly, but that is a price I am willing to pay. Transparency in reviews as well as in results and publications is worth it.

  4. SimonW says:

    I haven’t been paying enough attention, but my understanding was the paper denies the prior interpretation of the specific data, and that it doesn’t as such impact on “inflation theories” in general, just that this specific data is no longer considered evidence for inflation. Steve’s headlining seems to go further than this, but that may be unintentional emphasis.

    I’ve been a skeptic of inflation theories, and lots of cosmology, most of my life. A lot of cosmology is based on assumptions about the nature of physical laws (heterogeneity of spacetime as an example), and too little data. This is not cosmologists fault, data is hard to gather, and the interpretation has to be in the light of some of those assumptions (at least till we have the technology to test if those assumptions hold on none trivial amounts of spacetime), but I suspect too much weight is given to cosmological ideas which are probably more tentative than most physicists like to admit.

    Other evidence gathered over my life has supported inflation, so I’m MUCH less skeptical than I was, but I want good explanations of galactic formation, good understanding of the dark matter we can detect gravitationally, a good demonstration we know how much matter there is in our neighbourhood before I start assigning much weight to theories which are based on say the balance between expansion and the total mass of the Universe.

    Space is big as Douglas Adams mentioned, it seems likely it is getting bigger, and we need to account for it being the size it is at the apparent age it is, thus some sort of inflationary theory will likely win out, my skepticism is more to do with how we settled on the current ones. Even those hard told stories of theoretical physics, like theoretical prediction of relative abundance of elements, look a lot less convincing when you understand the maths and the assumptions made to produce the answer. If we oversell our certainty in these areas, science will be tarnished when results are overturned, I don’t think that has happened here, but it is one to be cautious of in discussing science. Results speak louder than words; a Smart Phone, a genetically modified yeast that makes human identical insulin, GPS, these are the stuff of science fiction from my childhood. Sure we want to understand the big picture, but people need to understanding where we are running, and where we are learning to crawl.

Leave a Reply