Inflation is dead, long live inflation! The very results hailed this year as demonstrating a consequence of inflationary models of the universe 鈥� and therefore pointing to the existence of multiverses 鈥� now seem to do the exact opposite. If the results can be trusted at all, they now suggest inflation is wrong, raising the possibility of cyclic universes that existed before the big bang.
In March experimentalists announced that primordial gravitational waves had been discovered. The team behind the had observed telltale twists and turns in the polarisation of the cosmic microwave background radiation (CMB) 鈥� the remnants of the earliest light produced in the universe.
Physicists thought the discovery was preliminary confirmation of inflation: the idea that for a sliver of a moment after the big bang there was a blisteringly fast expansion of the universe. The theory, the most widely held of cosmological ideas about the growth of our universe after the big bang, explains a number of mysteries, including why the universe is surprisingly flat and so smoothly distributed, or homogeneous.
Advertisement
But very quickly, the BICEP2 finding was shrouded in doubt, as it was revealed that the polarisation pattern could have been caused by cosmic dust. Cosmologists are waiting for space-based Planck telescope to reveal whether the dust could really make that pattern, and preliminary results released last week suggest dust might be able to.
But this week a team of theorists decided to analyse the polarisation signature further and ask: if it isn鈥檛 completely caused by dust, ?
Rather than just looking at the polarisation to see if it suggests the existence of gravitational waves, in Australia and his colleagues decided to look at the nature of those apparent gravitational waves to see if they were the type of waves predicted by inflation. And they weren鈥檛.
Counter to what the BICEP2 collaboration said initially, Parkinson鈥檚 analysis suggests the BICEP2 results actually rule out any reasonable form of inflationary theory.
Ruled out 鈥� possibly
Most inflationary models require that as you look at larger and larger scales of the universe, you should see stronger and stronger gravitational waves. Cosmologists call that a 鈥済ravitational wave spectrum鈥�.
鈥淲hat inflation predicted was actually the reverse of what we found,鈥� says Parkinson. How many inflationary models does it rule out? 鈥淢ost of them, to be honest.鈥�
It鈥檚 not entirely impossible for inflation models to conform with what BICEP2 found if you do 鈥渞eally tricky鈥� things to the mathematics, says , a theoretical astrophysicist at the University of Melbourne in Australia. But such tricky things would break something called the inflation consistency relation, she says, which is 鈥渟omething that鈥檚 considered pretty basic for inflation鈥�. The inflation consistency relation links the amplitude of the primordial gravitational waves with the distribution of matter in the universe.
, a physicist at the Massachusetts Institute of Technology who pioneered the concept of inflation, says the new analysis is convincing, but not so convincing that he鈥檚 ready to give up on the possibility that BICEP2 has a signal of inflation in it. The paper by Parkinson and his colleagues is 鈥渃ertainly a negative indication for a signal鈥�, he says, 鈥渂ut I will still reserve judgement until the joint analysis is released.鈥�
Guth thinks there could still be a signal that supports simple inflationary models. And he emphasises that if the signal does end up being dust, that is not evidence against inflation, since most inflationary models predict a much smaller signal that would require more work to find. 鈥淚f BICEP2 has not seen [evidence of] gravitational waves, then only certain inflationary models are ruled out, while the concept of inflation remains completely healthy.鈥�
at the Australian National University in Canberra, who has been critical of the theory of inflation, says he expects that further analysis will confirm that no gravitational waves were observed at all. 鈥淏ut on the other hand, if BICEP2 is shown to be correct, it鈥檚 exciting,鈥� says Schmidt. 鈥淎nd it does potentially break standard inflation and therefore you are testing inflation and showing its wrong.鈥�
, who helped develop inflationary theory but is now scathing of it, says this is potentially a blow for the theory, but that it pales in significance with inflation鈥檚 other problems.
Meet the multiverse
Steinhardt says the idea that inflationary theory produces any observable predictions at all 鈥� even those potentially tested by BICEP2 鈥� is based on a simplification of the theory that simply does not hold true.
鈥淭he deeper problem is that once inflation starts, it doesn鈥檛 end the way these simplistic calculations suggest,鈥� he says. 鈥淚nstead, due to quantum physics it leads to a multiverse where the universe breaks up into an infinite number of patches. The patches explore all conceivable properties as you go from patch to patch. So that means it doesn鈥檛 make any sense to say what inflation predicts, except to say it predicts everything. If it鈥檚 physically possible, then it happens in the multiverse someplace.鈥�
Steinhardt says the point of inflation was to explain a remarkably simple universe. 鈥淪o the last thing in the world you should be doing is introducing a multiverse of possibilities to explain such a simple thing,鈥� he says. 鈥淚 think it鈥檚 telling us in the clearest possible terms that we should be able to understand this and when we understand it it鈥檚 going to come in a model that is extremely simple and compelling. And we thought inflation was it 鈥� but it isn鈥檛.鈥�
Bouncing universe
Steinhardt favours newer theories that don鈥檛 require inflation to smooth out the universe. Instead of relying on inflation, which would produce big gravitational waves in the CMB, Steinhardt suggests the universe might have existed before the big bang, and slowly collapsed in a big crunch, before bouncing back and expanding anew. He thinks that could explain the smoothness of the universe, without invoking multiverses. Not finding gravitational waves in the years to come will be the start of evidence for this theory, he says. Other observable predictions are being developed but it鈥檚 a relatively new theory and more work is needed.
Schmidt has also been critical of inflation but is more ambivalent. 鈥淚t may well be that inflation does lead to a multiverse, but I would also say that I鈥檓 never sure of what we can and cannot predict. Ultimately it comes down to maybe there are things that are hard to know 鈥� like the multiverse. We may be in a shroud of ambiguity. But maybe not 鈥� we鈥檙e very, very good at coming up with things we never thought of before.鈥�
The next step is to see what the Planck data 鈥� due in the next month 鈥� say about the exact nature of cosmic dust. With BICEP2 in place and several new instruments on their way, all the cosmologists New 杏吧原创 spoke to say it is an exciting time.
Journal reference:
