Commentary: Time to nail inflation

  • 02 February 2008
  • From New Scientist Print Edition. 
  • Lawrence Krauss


IN THE beginning, there was inflation, or so we are told. I am not referring to American politics or the economy, but to a seemingly less contentious arena: the early universe.

Inflation - a brief period during which the universe expanded faster than the speed of light - remains the dominant paradigm of cosmology. And for good reason: it explains otherwise anomalous features of our universe, including the existence of stars and galaxies. Tiny ripples of quantum uncertainty stretched out to astronomical proportions: what could be more poetic?

Inflationary models, however, are so robust and malleable that it sometimes seems they can be tweaked to fit any observations - and from a scientific perspective, that's a problem. To decide whether a theory bears physical truth, one must be able to falsify it; a theory that can accommodate any observation is not a scientific theory at all.

The theory can be tweaked to fit almost any observations - and that's a problem

Luckily, inflation does not quite fit the category of "unfalsifiable". If it had to be very fine-tuned to match observations, theorists would feel at the very least that it was extremely unlikely, if not ruled out. As things stand, observations of the pattern of galaxies throughout the sky and of the cosmic microwave background (CMB) radiation left over from the big bang all precisely match what the simplest inflationary ideas predict, suggesting that inflation is on the right track.

So we find ourselves at a crossroads. Inflation is theoretically very attractive and seems to fit observations, but we don't know for sure. Where do we go from here?

One way is to gather new data. If inflation really caused the universe to expand at extreme speeds, it would have sent shock waves rippling through the fabric of space-time itself. These ripples, known as gravity waves, should be observable today: their effects might be seen in the CMB as fluctuations in the light's polarisation. A detector capable of picking this out could show us inflation's smoking gun.

But what happens if we spend a lot of money building new CMB detectors and we don't see polarisation? That won't invalidate inflation altogether, it will just rule out the many inflationary models that predict an observable signal. Is that worth perhaps a billion dollars?

That was the dilemma facing the US National Academies panel when asked by NASA and the Department of Energy to review five possible experiments that comprise NASA's Beyond Einstein programme. The candidates included a satellite to probe the nature of dark energy, a gravitational wave detector called LISA, and a CMB "inflation probe" satellite.

The panel chose the dark energy satellite. The technology behind LISA is new, and the panel deemed it not to be ready. But why did the CMB inflation probe lose out to dark energy? The panel reasoned that even if the dark energy probe fails to tell us anything new about dark energy - which, in my view, is likely - the satellite will provide a lot of "ancillary science". The inflation probe, on the other hand, was seen as a risky all-or-nothing gamble: either we see polarisation or we don't, and if we don't, that won't rule out inflation.

I can only hope that this decision will not kill future efforts for a NASA mission to probe inflation in the next decade. Even if we see nothing, it will imply that many of the simplest inflationary ideas need to be modified. It might even - gulp! - mean that some ideas from string theory are correct. If we do see something, we will be able to directly probe particle physics at an energy scale well beyond what's accessible at accelerators but low enough that trusted physics still applies. Either way, it sounds pretty exciting to me.

Cosmology - Keep up with the latest ideas in our special report.

From issue 2641 of New Scientist magazine, 02 February 2008, page 48


Click here