Did the universe start with a bang or a bounce—or something else entirely? The question of our origins is one of the thorniest in physics, with few answers and lots of speculation and strong feelings. The most popular theory by far is inflation, the notion that the cosmos blew up in size in the first few fractions of a second after it was born in a bang. But an underdog idea posits that the birth of this universe was not actually the beginning—that an earlier version of spacetime had existed and contracted toward a “big crunch,” then flipped and started expanding into what we see today. Now a new study suggesting a twist on this “bounce” scenario has supporters excited and inflation proponents newly inflamed over a “rival” they say they have repeatedly disproved, only to have it keep bouncing back.

Inflation has many admirers because the rapid expansion it posits seems to explain numerous features of the universe, such as the fact that it appears relatively flat (as opposed to curved, on large scales) and roughly uniform in all directions (there is roughly the same amount of stuff everywhere, again on large scales). Both conditions result when regions of space that ended up very far away initially started out close together and in contact with one another. Yet the latest versions of the theory seem to suggest—even require—that inflation created not just our universe but an infinite landscape of universes in which every possible type of universe with every possible set of physical laws and characteristics formed somewhere. Some scientists like this implication because it could explain why our particular universe, with its seemingly random yet perfectly calibrated-to-life conditions, exists—if every type of cosmos is out there, it is no wonder that ours is, too. But other physicists find the multiverse idea repulsive, in part because if the theory predicts that every possibility will come to pass, it does not uniquely foretell a universe like the one we have.

“Big bounce” theories also predict a flat and uniform cosmos, thanks to smoothing-out effects on space that can take place during the contraction. But the sticking point of the bounce idea has long been the transition between shrinking and expanding, which seemed to require the much-hated idea of a “singularity”—a time when the universe was a single point of infinite density—seen by many as a mathematically meaningless proposition that indicates a theory has gone off the rails. Now physicists say they have found a way to calculate the bounce without encountering any singularities. “We found we could describe the quantum evolution of the universe exactly,” says study co-author Neil Turok, director of the Perimeter Institute for Theoretical Physics in Ontario. “We found that the universe passes smoothly through the singularity and out the other side. That was our hope, but we’d never really accomplished this before.” He and Steffen Gielen of Imperial College London published their calculations last month in Physical Review Letters.

A Quantum Cosmos

The breakthrough came thanks to two techniques the researchers adopted. One was to use the nascent and still-not-complete theory of quantum cosmology—a mash-up between quantum mechanics and general relativity—instead of classical general relativity to describe the universe. The second was to assume that when the cosmos was young matter behaved like light in that the laws of physics that describe it did not depend on scale. For example, light acts the same regardless of its wavelength. The physics of matter, on the other hand, usually vary from small to large scales. “We know that in the first 50,000 years the universe was essentially just filled with radiation,” says Anna Ijjas, a physicist at Princeton University who was not involved in the research. “The normal matter we see now was not really very significant. I think a scaleless early universe is actually very much suggested by our current measurements.”

Under those conditions Turok and Gielen found that the contracting universe would never actually become a singularity—essentially it would “tunnel through” the worrisome point by hopping from a state right before it to a state right after it. Although such sidestepping sounds like cheating, it is a proved phenomenon in quantum mechanics. Because particles do not exist in absolute states but rather hazes of probability there is a small but real chance they can “tunnel” through physical barriers to reach locations seemingly off-limits to them—the equivalent, on a microscopic scale, of walking through walls. “The fuzziness in space and time and the matter conspires to make it uncertain where the universe is at a given time,” Turok explains. “This allows the universe to pass through the singularity.”

Other big bounce proponents say the work is a significant step. “By making those two plausible assumptions, they find a very interesting result, which is that a bounce can occur,” says Princeton physicist Paul Steinhardt, one of the founders of inflation theory who has more recently become one of its sharpest critics. He was not involved in Turok and Gielen’s study. “It shows that in principle a singularity can be avoided.” Steinhardt and Ijjas have been working on another way to mathematically demonstrate the possibility of a bounce, by introducing to the universe a special type of field that causes the contraction to turn into expansion well before space gets small enough to become a singularity. Their solution uses classical general relativity as opposed to quantum cosmology. “It means that classical, nonsingular bounces are also possible,” Steinhardt says. They reported their work in a paper posted June 28 to the preprint server arXiv.org.

Both studies are still preliminary. Turok and Gielen were able to calculate the bounce only for the case of an idealized universe that is completely smooth and lacks the small density fluctuations that lead to the formation of stars and galaxies in the real cosmos. “The cases that we can actually solve exactly are very simple universes,” Gielen says. “The question you always have is, ‘Will that still be there if you go to something more complicated?’ That’s what we’re working on at the moment.”

If the universe bounced once, a natural question is whether it will again. But not all bounce theories suggest we are destined to cycle forever through contractions and expansions—for example, even if our universe bounced before, we have no indication so far that it is heading for another contraction. The dark energy thought to make up the largest chunk of the cosmos’ total mass–energy budget seems to be pulling our universe apart at an ever-accelerating rate. What is truly in store for the future is a very open question—about as open, in fact, as the issue of how it all got started.

Bad Blood

Many advocates of inflation are highly skeptical of any bounce model, especially because they say proponents had repeatedly claimed in the past to be able to calculate bounces without singularities, only to be disproved. “I’m not happy that they do not admit that all their earlier papers should be disregarded,” says Stanford University physicist Renata Kallosh, who calculated errors in previously proposed bounce models. “They now make a new claim, and this new claim I don’t believe.” Alan Guth, a pioneer of inflation based at Massachusetts Institute of Technology, agrees. “I’m still skeptical whether they have actually achieved a nonsingular solution,” he says. “I would like to wait and see how it develops. If they have succeeded in what they claim they’ve done, I do agree it’s very important—even if it’s not the best model for the history of the universe.”

Some inflation researchers are more forgiving, though. “I think that this is a very intriguing line of research,” says Marc Kamionkowski of Johns Hopkins University. “The bounce scenarios, although not yet developed to the level that inflation has been developed, are promising, and it’s imperative to try to develop them further. This paper provides an interesting mathematical result, in a toy model,” he adds, referring to the idealized universe the researchers worked with.

Kallosh and others object to using quantum cosmology to describe a bounce, because the universe may not have been microscopically small at such a stage. “They have the collapse of a big universe—why should a big universe be different from what general relativity says?” Turok counters that any ultimate theory of the universe will have to incorporate quantum mechanics into general relativity, because the classical theory on its own is known to break down at certain extremes. “Nature is quantum,” he says. “We know that classical theories don’t make any sense at a very basic level.”

Turok and other critics of inflation have their own problems with the dominant theory. They charge that inflation requires unlikely circumstances to get started (a claim proponents disagree with) and that it does not resolve the specter of a singularity at the moment of the big bang itself. Furthermore, “inflation leads to this nightmare scenario of a multiverse,” Turok says, “which for some strange reason is surprisingly popular.”

He suggests that the heated debate in the field and the heavy scrutiny new ideas receive will help scientists ultimately converge on a better theory of our origins. “People hold very strong opinions,” Turok says. “I freely admit I do and I freely admit my opinions aren’t shared by 95 percent of cosmologists. I’m actually critical of all these theories, including the ones I invented. But today we have spectacular observations pointing us at incredible simplicity in the universe. To me that means that all of our existing theories are way too complicated. The observations are pointing at simplicity and it’s our job to come up with a simple theory that will hopefully explain those.”