Default publications post thumbnail

Rethinking the Latest Evidence for Inflation

In 1964, two scientists serendipitously found a critical feature of our universe: the cosmic microwave background radiation (CMB). This discovery dethroned steady-state models and simultaneously established big bang cosmology as the proper understanding of how the universe began and developed. Yet the extreme uniformity of the CMB also posed a problem. How could all regions of the universe have the same temperature given how fast light travels? In the early 1980s, physicists Alan Guth and Andrei Linde “solved” that problem by postulating inflation, a period of extremely rapid expansion in the earliest fractions of a second after the universe’s birth. Then in 2014, almost 50 years after the CMB finding, scientists announced with great fanfare the discovery of “smoking gun” evidence supporting this inflationary model.

Now a recent publication shows that dust in the universe, and not inflation’s gravity waves, explains the evidence. What are we to make of this finding?

First, the correction of last year’s enthusiastic announcement serves as a reminder to beware of hype and to exercise caution regarding significant findings. (Fortunately, hype rarely affects the technical journals or science news sites written by science journalists, such as ScienceDaily.) In this instance, the hype did result in understating important caveats and missing concerns about the data analysis. Looking back at the article I wrote in April 2014, the tone conveys that finding inflation’s gravity waves was a done deal. Sure, it contained the obligatory if-affirmed-by-future-research statement, but my tone communicated far more certainty—mainly because I expected (and wanted) evidence affirming the inflationary model.

Second, even though the gravity wave evidence has turned out to be inconclusive, there remains a wealth of experimental data supporting inflationary cosmology. Inflation’s predictions include

  • a flat universe—predicted when scientists could find only 25–30 percent of the energy density required by a flat universe;
  • adiabatic density fluctuations—which gives three or more peaks in the multipole distribution of the CMB; and
  • near-scale invariance of density peaks—Planck finds ns = 0.9603 ± 0.0073, near but not exact scale invariance, as expected.1

In fact, the latest findings about the gravity waves bring the BICEP2 and Planck levelsin line with one another.2

Third, the whole process highlights the integrity of the scientific process. When the first announcement broke, many scientists were excited about the smoking gun signal of inflation. Others expressed skepticism. Deeper investigation revealed that the data was not sufficient at the time to make a final determination. The BICEP2 team and the Planck team both put forth a concerted effort to acquire the necessary data and perform the proper analysis. In the end, all sides agree that at this time no conclusive evidence exists for the gravity waves that inflation models predict. Ultimately, the data—not a desired outcome—determined the conclusion that scientists drew.

One final point warrants mention. The BICEP2 and Planck data show no conclusive evidence for gravity waves. However, the data do indicate some presence of the gravity wave influence, but at a level too small to claim a significant detection. This means that increased sensitivity may detect the gravity wave in the future.

Scientific progress often happens with both advances and setbacks. The sequence of events over the past year demonstrates that we need to watch out for unwarranted hype, acknowledge the wealth of data supporting inflation, and recognize the integrity of the scientific progress. And don’t forget that the detection of inflation’s gravity waves might be just around the corner.

  1. Planck Collaboration, “Planck 2013 Results. XXII: Constraints on Inflation,” Astronomy and Astrophysics 571 (November 2014): id. A22.
  2. The BICEP2/Keck and Planck Collaborations, “A Joint Analysis of BICEP2/Keck Array and Planck Data,” provisionally accepted by Physical Review Letters, 2015,