Could there be other universes beyond ours?

What do scientists mean by other universes?

In a 2003 survey, Max Tegmark laid out the four levels described above. He argued that the key question is not whether parallel universes exist but how many levels there are. In his view, the first level, distant regions of our own space, is uncontroversial in the standard picture of cosmology. He also argued that the higher levels face a severe “measure problem” before they can make testable predictions. Others are more cautious, including the authors of the Stanford Encyclopedia’s cosmology entry, discussed next.

Level one depends on how big space is. The part we can see is limited, because light has had only so much time to travel. Whether space beyond it goes on forever is unknown. Our answer Is the universe infinite? explains why. Tegmark argues that if space is infinite in this way, there would even be an identical copy of you, unimaginably far away.

What are bubble universes?

Cosmic inflation proposes a brief burst of very fast expansion in the universe’s first instant. The Stanford Encyclopedia’s fine-tuning entry notes that its strongest support is widely believed to come from its apparently correct predictions for the pattern in the Big Bang’s afterglow. For more on that instant, see What Happened in the First Second After the Big Bang?

The Stanford Encyclopedia’s cosmology entry, by Christopher Smeenk and George Ellis, explains that many cosmologists argue for a version called eternal inflation. In some regions inflation stops, forming “pocket” universes. In others it goes on. Supporters argue that we should accept the multiverse the way we accept many things we cannot see directly: as an unavoidable consequence of an established physical theory.

The same entry lists strong doubts. Not every inflation model leads to eternal inflation, arguably including the models the afterglow data favor, and the accounts of how it happens rely on speculative physics. The idea that each pocket gets different constants is an assumption, not yet backed by a well-tested theory. And Alan Guth, as the entry quotes him, said that in eternal inflation “anything that can happen will happen.” If everything happens somewhere, the entry warns, then any observation fits the theory.

Could the multiverse ever be tested?

Scientists have tried. One idea is that a neighboring bubble might once have bumped into ours, leaving a round scar in the cosmic microwave background, the faint afterglow of the Big Bang. In 2011, Stephen Feeney and three colleagues searched seven years of data from NASA’s WMAP satellite. They found that the data did not call for bubble collisions. They set a limit, at a 68% confidence level: fewer than 1.6 detectable collisions on the whole sky, on average. They added that data from the Planck satellite could test the idea more fully.

A missing scar does not rule out the multiverse. The Stanford Encyclopedia notes that the theory does not always predict scars; a collision that happened too early would be erased.

The second approach is to predict the constants of nature. In 1987, Steven Weinberg reasoned about which values of the energy of empty space would still let galaxies form, in a multiverse where that value varies. According to the Stanford Encyclopedia, it is the most widely cited evidence for a multiverse, and some advocates see its success as the most compelling evidence. Others point to rival explanations for the same value.

The strongest objection. The Stanford Encyclopedia notes the unease many feel about multiverse ideas, and the “measure problem”: nobody agrees how to count probabilities across the ensemble. In 2014, George Ellis and Joe Silk argued in Nature that attempts to exempt speculative theories of the universe from experimental verification undermine science. Supporters reply, as the Stanford Encyclopedia puts their case, that we accept many things we cannot see directly when they follow from an established theory. The debate continues.

What about the many worlds of quantum physics?

The many-worlds interpretation, going back to Everett in 1957, says that every possible outcome of a quantum measurement really happens, each in a different world. We see only one.

Supporters like it mainly because it avoids the sudden “collapse” to one outcome that other readings need. On this view, there are many versions of you in different worlds. In the Stanford Encyclopedia’s words, it allows nature to avoid “playing dice.” (For Einstein’s famous doubt about dice, see Did Einstein lose his argument that God does not play dice?)

Can it be tested? The entry says a decisive test would need interference between states of a large, everyday-sized object, which is far beyond today’s technology. Some rival theories that make the wave collapse have been tested, and some versions were ruled out, but not all. Realistic experiments, the entry adds, cannot tell many worlds apart from Bohmian mechanics, a reading with a single world. The entry’s author, Lev Vaidman, argues that the burden of proof lies with opponents, since collapse would be new physics beyond the well-tested equations of quantum theory.

How is a braneworld different from the movie multiverse?

In films, other universes usually hold other versions of people you can visit. A scientific braneworld is a different kind of idea. In a 1999 paper, Lisa Randall and Raman Sundrum studied a single three-dimensional brane, or sheet, inside a space with five dimensions. They showed that ordinary gravity could still come out right.

Other physicists went further. The ekpyrotic model of Justin Khoury and colleagues proposed that the hot Big Bang came from a brane colliding with another boundary in the larger space. They pointed to a way to tell it apart from inflation, using its pattern of gravitational waves, ripples in space itself.

CERN explains how experiments at the Large Hadron Collider look for signs of extra dimensions: heavier copies of known particles, energy that goes missing, or tiny black holes. So the difference is this. A scientific braneworld comes with equations and tests about gravity, particles and ripples in space. None of these papers is about traveling to another world.

What do people often get wrong?

“Scientists have found other universes.” They have not. No observation so far has shown one.

“The multiverse is pure science fiction.” Several versions grow out of serious theories, and some have been put to real tests with real data.

“A multiverse settles the God question.” Many see it as the main rival to design as an answer to fine-tuning. But as our page Is the universe fine-tuned for life? explains, critics such as Ian Hacking argue that inferring other universes from our lucky one is a mistake. For the design side, see What are the main arguments for God’s existence?

“Other universes would have different physics.” Not always. Tegmark’s first level shares our laws. For the hope of one theory behind all laws, see What is a theory of everything?

How can you check this yourself?

• Ask which multiverse. Far-off space, bubbles, many worlds and branes are different claims with different evidence.
• Ask what would count against it. A good claim names an observation that could go the other way.
• Read the abstracts. The Tegmark survey, the Feeney search and the Randall–Sundrum paper are free on arXiv.
• Watch for “proves.” A headline that says scientists proved or found a parallel universe goes beyond what any study so far has shown.

The answer stays the same after all this detail. Other universes are allowed by several serious theories, none has been observed, and whether most of them can be tested is still debated.