Time and the Multiverse

You must remember this, a kiss is still a kiss, a sigh is just a sigh. The fundamental things apply, as time goes by...

For many of us, this song from Casablanca evokes memories of romance. But to Laura Mersini-Houghton it may provoke deeper questions about the foundations of reality: Why can we only remember the past, not the future? And what is the origin of the arrow of time? Her conclusion is that the time’s arrow is not as fundamental as we may believe.

Mersini-Houghton’s view of time comes from looking at the bigger picture—the biggest possible picture, in fact. With the help of a $50,000 grant from FQXi, she is investigating the possibility that our universe is just one of many in a "multiverse" of universes and what that means for time’s arrow.

Womb to Tomb

We travel through life from womb to tomb, not vice-versa, yet physicists have no real explanation for why time flows in only one direction. The microscopic laws that underlie the behavior of particles have no such arrow, working equally well backwards as forwards. So why doesn’t time run backwards?

Physicists usually explain the arrow of time using the concept of increasing entropy—a measure of the disorder of a system. The universe evolves from a highly-ordered, low-entropy beginning, to a progressively more disordered state, defining time’s arrow. So, sugar cubes dropped into coffee dissolve as time passes, increasing the disorder of the coffee-sugar system; but they do not re-solidify. To Mersini-Houghton, a physicist at the University of North Carolina, Chapel Hill, however, this reasoning simply begs the following question: Why did the universe begin in a highly unlikely low-entropy state in the first place?

The multiverse view offers a simple answer: If there are an infinite number of cosmoses, it would make sense that at least some universes should start in a low entropy state. Mersini-Houghton first became interested in the idea of a multiverse with the advent of the string-theory landscape. In 2003, string theorists began to realize that their equations offered a staggering 10⁵⁰⁰ equally-valid solutions, each of which could describe a possible universe. Suddenly there was talk of a string landscape—a multiverse of universes, each with different physical laws.

This landscape provides the basis of Mersini-Houghton’s model. Each new universe forms in this landscape as a bubble, with some intrinsic vacuum energy—similar in character to the dark energy causing today’s universe to expand at an ever-increasing rate—that drives it to inflate. At the same time, matter causes the bubble to try to crunch back down. "It is this tug-of-war that determines whether a universe can be born or not," says Mersini-Houghton.

Crucially, in this model, only high-energy bubbles will overcome the matter crunch and grow into full-blown universes. These bubbles would have started out with low entropy, explaining the seemingly unlikely initial conditions of our universe.

Arrowless Time?

Mersini-Houghton claims that time in the larger multiverse is arrowless, favoring no particular direction. But while the physical laws in each baby universe inherit this time-symmetry, the bubbles themselves do not. That’s because at the moment of its birth, the bubble loses information about the multiverse. This information loss results in growing disorder, says Mersini-Houghton. In other words, entropy starts to increase in the bubble, creating a local arrow of time.

To describe the dynamical evolution of the bubble, Mersini-Houghton uses a quantum-mechanical master equation. "It’s important to derive everything from an existing, fundamental theory," she explains. But this comes at a price: Since there is currently no working theory that unites quantum mechanics with general relativity, Mersini-Houghton assumes that the laws of quantum mechanics are more fundamental than those of general relativity.

However, not all cosmologists think the role of general relativity can be downplayed. FQXi essay contest winner Sean Carroll at Caltech, in Pasadena, is also independently tackling the issue of time in the multiverse but, by contrast, he is using a semiclassical quantum-gravity approach. His model also contains a parent multiverse, with no overall arrow of time.

"You start with a spacetime that has no directionality of time, which makes sense given that the underlying physical laws are like that," says Carroll. In his model, baby universes preferentially begin with low entropy. Once grown, each universe can eventually give birth to new universes, which arise due to quantum fluctuations in the presence of vacuum energy. These universes have arrows of time pointing in different directions, so, in some universes, time could actually run backwards.

Time flowing backwards is less mind-boggling than it sounds, says Carroll reassuringly. In what we call the past there are universes where, from our point of view, entropy is increasing towards the past. But in the same way, from their point of view, entropy is increasing towards the future and we’re in their past. "Everything is completely symmetric," he says. "Backwards is in the eye of the beholder."

This all sounds nice in theory. But given that we are locked within our universe, can we possibly ever test whether a multiverse exists, let alone whether it has an arrow of time?

Mersini-Houghton thinks so. In her model, baby universes retain a vague, gravitational influence on each other even after they lose information about the multiverse. In 2006, she predicted that this ghostly cross-talk would result in a peculiar ‘pull’ on galaxy clusters in one part of the sky. Last year, such an effect, dubbed "dark flow," was surprisingly observed by NASA. There is no consensus as to what causes the flow, but Mersini-Houghton is excited: "I never thought we’d see anything like this, at least not in my life time."

Mersini-Houghton had also predicted that a neighboring universe could create a giant void in space by pushing against us, repelling gravity and matter. A few months later, the discovery of such a void using radio telescopes at the Very Long Baseline Array in New Mexico was reported. But this result was later called into question in 2009, when astrophysicists at the University of Michigan claimed that the evidence for the void was a statistical blip.

Despite this setback, Rüdiger Vaas, a philosopher of science from the University of Stuttgart, says that one of the main advantages of Mersini-Houghton’s model is that it is testable. "You can derive predictions and it is possible to test them right now," he says. "Although it’s currently not clear whether it’s true or false, it looks very promising."

So will Mersini-Houghton’s latest predictions ultimately stand up to scrutiny? Only time will tell...