Theoretical physicist Brian Greene explores the possibility that our universe is not the only one, arguing that numerous independent developments in fundamental physics point toward some version of a multiverse: a vast collection of parallel universes. He traces nine distinct multiverse proposals, each emerging from a different branch of physics, and examines whether such ideas can legitimately be called science.
Greene begins with the Quilted Multiverse, a proposal rooted in the idea that if space extends infinitely, the finite number of possible particle arrangements within any observable region must eventually repeat across the cosmos. He provides background on general relativity, Albert Einstein's 1915 theory reconceiving gravity as the curvature of spacetime caused by matter and energy, and recounts how physicist Georges Lemaître and mathematician Alexander Friedmann independently found that Einstein's equations implied an expanding universe, a conclusion Einstein resisted until astronomer Edwin Hubble's 1929 observations confirmed it. Greene explains that quantum mechanics limits distinguishable particle configurations within any observable region to roughly 10 raised to the power of 10^122. In an infinite cosmos containing infinitely many such regions, these finite arrangements must repeat, guaranteeing the existence of exact replicas of our observable universe somewhere in the distant reaches of space.
The second multiverse arises from inflationary cosmology. Standard big bang theory faces the horizon problem: The cosmic microwave background radiation, a sea of photons from the universe's earliest moments, is uniform to extraordinary precision across regions that have never been in contact. Physicist Alan Guth proposed a resolution, refined by physicists Andrei Linde, Paul Steinhardt, and Andreas Albrecht, positing a brief burst of fast expansion in the universe's first fraction of a second, driven by the repulsive gravity of a hypothetical entity called the inflaton field. Quantum fluctuations cause the inflaton field to drop from its high-energy state at different times in different places, creating bubble universes where inflation has ended and particles have formed. The surrounding space continues inflating faster than new bubbles form, ensuring the process never stops. Different bubbles can have different physical properties because fundamental fields can settle into different values in each. Each bubble, though finite when viewed from outside, appears spatially infinite to its inhabitants, meaning each harbors its own Quilted Multiverse.
Greene then turns to string theory, which proposes that every fundamental particle is a tiny vibrating filament of energy, with different vibrational patterns producing different particle properties. The theory requires six extra spatial dimensions curled up so small as to evade detection. Their shape, drawn from a class called Calabi-Yau shapes, determines particle properties, but string theory has not identified which shape corresponds to our universe. In the mid-1990s, physicists discovered that string theory also contains higher-dimensional objects called branes (short for membranes). If our universe exists on a three-dimensional brane, other branes floating nearby in the higher-dimensional space would constitute parallel universes, forming the Brane Multiverse. Ordinary matter and light remain confined to our brane; only gravity can travel between branes. Steinhardt, physicist Neil Turok, and collaborators proposed that collisions between branes could produce rushes of particles resembling new big bangs, with repeated collisions generating universes parallel not in space but in time. This Cyclic Multiverse avoids the need for a cosmic beginning.
The discovery that fields called brane fluxes can thread through the Calabi-Yau shapes increased the number of possible forms for the extra dimensions to roughly 10^500. This became significant when, in the late 1990s, astronomers studying distant supernovae discovered that cosmic expansion has been accelerating for the past 7 billion years, implying that the cosmological constant, a measure of the energy intrinsic to space, has a tiny but nonzero value of about 10^−123 in Planck units. This value is more than 120 orders of magnitude smaller than quantum field theory predicts. In 1987, physicist Steven Weinberg proposed that if the cosmological constant varies across a multiverse, we necessarily inhabit a universe where the value is small enough to permit galaxy formation. Through eternal inflation, quantum tunneling events (spontaneous jumps between energy states) create nested bubble universes realizing different configurations from string theory's vast landscape of possibilities. Greene calls this the Landscape Multiverse.
These proposals raise a pointed question: Is invoking unobservable parallel universes legitimate science? Greene argues that science has long relied on inaccessible constructs, from electromagnetic fields to quantum probability waves, and that confidence in them derives from the predictions they enable. A multiverse theory could yield testable predictions if physical features are distributed in sharply skewed or correlated ways across its universes. However, formidable obstacles remain, including the measure problem: the difficulty of comparing infinite collections of universes when no agreed-upon method exists for assigning relative sizes to infinite sets.
The Quantum Multiverse emerges from physicist Hugh Everett III's challenge to the Copenhagen interpretation, the standard view that measurement causes a particle's probability wave to collapse to a single outcome. Everett, a Princeton graduate student in the 1950s, proposed taking the Schrödinger equation at face value: Measurement does not collapse the wave but causes it to evolve into separate branches, each containing a copy of the observer registering a different result. Through decoherence, the involvement of macroscopically many particles causes the branches to become effectively invisible to one another, so every quantum possibility is realized in its own branch of reality. Greene notes that the most serious challenge to this Many Worlds interpretation is explaining how probability retains meaning when all outcomes occur.
The Holographic Multiverse arises from theoretical studies of black holes. Physicist Jacob Bekenstein proposed that black holes carry entropy, a measure of hidden information, proportional to the area of their event horizon, the boundary beyond which nothing can escape. Physicist Stephen Hawking's 1974 calculation confirmed this by showing that black holes emit thermal radiation. The key surprise is that information storage capacity scales with surface area rather than volume. Physicists Gerard 't Hooft and Leonard Susskind generalized this into the holographic principle: All physical phenomena within a region of space can be fully encoded on its bounding surface. In 1997, string theorist Juan Maldacena provided a concrete realization, establishing that a gravitational string theory in the interior of a particular curved spacetime is physically equivalent to a non-gravitational quantum field theory on its boundary. This duality forms the basis of the Holographic Multiverse: Two radically different physical frameworks describe the same reality, suggesting our three-dimensional experience may be a holographic projection from a distant two-dimensional surface.
Greene then considers more speculative proposals. If consciousness arises from information processing rather than any specific biological substrate, sufficiently advanced computers could generate simulated worlds with self-aware inhabitants. Philosopher Nick Bostrom argues that if such simulations are possible, simulated beings would vastly outnumber biological ones, making it statistically likely that we ourselves are simulated. This is the Simulated Multiverse. The Ultimate Multiverse, championed by physicist Max Tegmark, proposes that every possible mathematical structure corresponds to a physically real universe. Greene expresses skepticism about this most expansive proposal, noting that unlike the others it lacks a physical process that could generate it.
In his conclusion, Greene observes that the multiverse concept may complete the Copernican revolution by demoting our entire universe from any central position. Multiverse theories collectively challenge traditional science by rendering three mysterious features into environmental facts that vary from universe to universe: initial conditions, constants of nature, and mathematical laws. Greene closes by reflecting on the power and limits of mathematics, acknowledging that whether the equations leading to parallel universes are being taken too seriously or not seriously enough is a question only future investigation can answer.