Endless universes mean empty skies

Two popular philosophical questions are:

• Where are all the aliens? (the Fermi paradox)
• Are we in a simulation? (the simulation hypothesis)

I posit that there is a single answer to both of these questions that is currently being overlooked.

The Fermi Paradox

There are 300 billion stars in our galaxy and a trillion planets, providing ample space for life to develop. Our universe is 14 billion years old, providing ample time for alien life to reach technological maturity. One could colonize the entire galaxy in 10 million years--a mere instant on astronomical timescales--with von Neumann probes traveling at 1% the speed of light. So when the SETI telescopes scour the night sky, why don't we see evidence of alien life. Where are the Dyson swarms and other extraterrestrial megastructures? ^[aside]

There are many proposed solutions to the paradox. A common one is that there must be some kind of "Great Filter" that prevents nearly 100% of planets from producing intelligent life capable of spreading across the galaxy.

For example, maybe abiogensis--the moment geochemistry became biochemistry--is the Great Filter? Meaning that the chance of life occurring at all is essentially zero. However, life appeared on Earth remarkably quickly after the planet cooled. This rapid emergence suggests that given suitable conditions, life might not be extraordinarily improbable. ^[aside]

^[aside]

Maybe the leap to complex/intelligent/technologically-advanced life is the Great Filter? But multicellularity evolved independently multiple times. And any environment complex enough to support life creates selection pressures for prediction, learning, and behavioral flexibility. Tool-use, problem-solving, and social cooperation have evolved in many species on Earth with a vast phylogenetic distances between them. Multiple human civilizations independently developed writing, mathematics, and metallurgy. Technological progress has followed the same exponential trajectory that we see in all of Earth's history--from prokaryotes to eukaryotes to multicellularity to intelligence, each leap enabling the next in less time. Life appears to be driven by a fundamental process that exponentially increases negentropy--inexorably marching toward greater complexity as it captures and channels free energy at an accelerating pace.

Maybe the Great Filter is ahead of us? Perhaps advanced civilizations destroy themselves before reaching galactic scales. But inevitable self-destruction would require remarkably strong convergence among vastly different lifeforms in vastly different environments. Alien psychologies, social structures, and histories will of course be much different from ours. A hive-mind species might never develop nuclear weapons due to lack of internal conflict and would probably be very careful about developing any potentially dangerous technologies. We've already survived 80 years with nuclear weapons, suggesting at least some civilizations could navigate dangerous technologies. ^[aside]

There are many more proposed solutions ^[aside] but each one faces substantial counterarguments, which is precisely why the Fermi paradox remains so perplexing.

The Simulation Hypothesis

The question of whether the entire universe is simulated is an entirely separate question from Nick Bostrom's simulation hypothesis. For our whole universe to be simulated, the real universe would have to be unfathomably more complex than ours in order to have the computing power necessary to achieve this, like how I can use my computer in our reality to simulate a simpler Minecraft world. Maybe we're inside Russian nesting dolls of increasingly simpler simulated universes, but this is outside the scope of this essay. The simulation we are talking about is kind that a technologically-mature civilization in our universe would be capable of creating, more akin to an extremely immersive video game than a full universe simulation.

Bostrom in his 2003 paper "Are You Living in a Computer Simulation?" argues that at least one of three propositions must be true: 1. Civilizations at our level of development almost never reach technological maturity. They go extinct before developing the computing power necessary to run ancestor simulations ^[aside]. 2. Advanced civilizations almost never run ancestor simulations. 3. We are almost certainly living in a simulation. If advanced civilizations do run many ancestor simulations, then most beings with our kinds of experiences are simulated rather than biological originals.

The argument is quite simple:

• If propositions 1 and 2 are false (civilizations do reach maturity and they do run simulations), then there must be many simulations.
• If there are many simulations, we're probably in one.

I think the most common type of simulation in a technologically-mature, post-intelligence-explosion civilization would be for entertainment purposes. You could transfer your brain from a biological substrate to an electrical one (one neuron at a time à la the Ship of Theseus to avoid breaking your stream of consciousness and avoid raising philosophical questions about what is actually "you"); or alternatively, become a "brain in a vat". Then, you could have a digital superintelligence construct a fun simulated environment for you and your loved ones, and feed that artificial sensory information into your brain. And of course, to make it as immersive as possible, temporarily remove your prior memories such that you aren't aware you're in a simulation. A being with functional immortality would probably spend quite a lot of time in these simulations. If I found myself among the world's billionaires, I'd find the simulation hypothesis nearly irresistible. The sheer improbability of ending up as one of the most influential people during perhaps the most crucial moment in human history would strain credulity. But I hope that any powerful people that believe this would avoid solipsism and act with moral seriousness.^[aside]

The Self-Sampling Assumption

The statement "if there are many simulations, we're probably in one" is implied by the self-sampling assumption: that we should reason as if we are randomly selected from the set of all observers (past, present, and future) in our reference class.

Basically, if advanced civilizations run many ancestor simulations, then there would be vastly more simulated beings than original biological beings. For example, if one original civilization eventually runs millions of simulations of civilizations at our stage, then there are millions of simulated beings for every one "real" being. If you're a random sample from all beings with experiences like yours, and 99.99% of such beings are simulated, then you should conclude there's a 99.99% chance you're simulated.

The self-sampling assumption connects directly to the Fermi paradox. One might propose that the reason we don't see evidence of alien life is that we are just early. The argument is that during the tens of trillions of years over which stellar evolution will continue, advanced civilizations will colonize the galaxy, but we don't see them yet because we're among the first ones. This explanation faces a probabilistic objection. What is the chance of being early when the universe will exist for 10^100 years? It's akin to saying that all three propositions in Bostrom's argument are false--that the universe will be filled with simulations but we're real because we're extraordinarily lucky. Surely this is the least likely solution.

However, this is my argument. We're in base reality and we don't see alien life because we're the first civilization in the universe. Not in spite of the self-sampling assumption, but because of it. There's a subtle error in how we define the set of observers that, once corrected, reveals why we must necessarily find ourselves at the beginning of cosmic history, in base reality, seeing an empty universe.

To see this error, we must first examine how eternal inflation radically changes the set of all observers.

Cosmic Inflation

The universe almost certainly did not explode from a singular point. This is a common misconception. These days, the best accepted description of the time before the Big Bang is given by inflation theory--specifically, slow-roll inflation. Here is a simplified explanation of inflation:

The fabric of space is expanding slowly (AKA Hubble expansion).

• According to Einstein's equations, the fabric of space will stretch and expand when space itself has energy. The rate of expansion scales with the strength of the vacuum energy density.
• We can observe this phenomenon in the redshifted light from distant galaxies, suggesting that they are moving away from Earth at speeds proportional to their distance, rather than slowly moving towards us as you would expect from gravity alone. Imagine galaxies as ants on the surface of an inflating balloon.
• Hubble expansion is extremely slow. It's only noticeable to us when it adds up over vast distances, like between us and distant galaxies.

Space used to expand mindbogglingly rapidly (AKA cosmic inflation).

• Prior to the Big Bang and in a tiny fraction of a second after it, the fabric of space itself was in a rapidly inflating state, so fast that the radius of the universe increased by a factor of at least 10^26 in less than 10^-32 seconds.
• During this time, the universe had an incredibly high vacuum energy density.

The transition from fast to slow expansion is explained by quantum field theory (QFT).

• The universe is filled with quantum fields. A field is just some property that has a numerical value at every point in space, which we call the field strength.
• An elementary particle is just an oscillation in this field strength--a little packet of energy held by the field, a local vibration.
• A quantum field can contain intrinsic energy even without particles. This gives us the vacuum energy density needed for expansion/inflation.
• Quantum fields want to drop to the lowest energy state, converting intrinsic energy into particles. A quantum field with high intrinsic energy can be thought of as having high potential energy (like a ball sitting at the top of a hill).
• During the early universe, the "inflaton field"^[aside] was in a high energy state (high potential energy), giving us the high vacuum energy density needed for rapid inflation. Then, it converted its potential energy into the particles of the Big Bang (the ball rolled down the hill).
• Afterwards, the inflaton field settled in a local energy minimum (the ball rolled to the bottom of a valley), giving us the tiny constant vacuum energy density of Hubble expansion that we observe today.

Nearly every physicist believes the theory of cosmic inflation is true. It very neatly solves various conundrums in cosmology like the flatness problem, the horizon problem, and the magnetic monopole problem; it fits extremely nicely into our modern understanding of quantum mechanics and general relativity; and it predicts the patterns we observe in the Cosmic Microwave Background radiation (CMB).

The reason I'm bringing this subject into a discussion about the self-sampling assumption is that slow-roll inflation has a stunning and likely unavoidable implication: cosmic inflation, once started, is probably eternal, generating a multiverse of infinite bubble universes.

Eternal Inflation

Quantum fields fluctuate due to the intrinsic randomness of the quantum world. As the inflaton field rolls down the potential energy hill, the field strength should fluctuate slightly. This means some regions of the universe would finish inflation a little ahead of others, and that would lead to density/temperature fluctuations in the matter produced by the end of inflation. We see those fluctuations in the CMB. In fact, this is the best evidence we have in favor of inflation.

These fluctuations also suggest eternal inflation and multiple universes. Fluctuations back up the potential energy slope would cause inflation to last much longer in that spot. Remember that during inflation, the volume of space is multiplying by a factor of around 10^2500 every second^[aside], so an uphill fluctuation in a tiny patch of space would very quickly outgrow its surroundings, producing a new inflating region. That region would continue to decay spawning new universes, but also spawning new inflating regions.

The result is that inflation never stops, but rather forms a fractal structure of infinitely expanding space interspersed with bubble universes. And to get this entire process started you just need a tiny speck--a fraction of the Planck energy within a Planck volume. That speck will start expanding exponentially and become infinite universes.

Remember this from the previous section?: "[after inflation ended], the inflaton field settled in a local energy minimum (the ball rolled to the bottom of a valley), giving us the tiny constant vacuum energy density of Hubble expansion that we observe today."

Why would our universe get stuck in a false vacuum state with precisely the vacuum energy density needed for stars and galaxies to form? Our current understanding of QFT predicts that the "cosmological constant" (our universe's vacuum energy density ) should probably be around 10^120 times larger than it is, creating expansion so violent that not even subatomic particles could stay bound together.

This is part of a broader question known as the finetuning problem. Why do so many aspects of physics seem finetuned for a universe that can support life? Setting aside design arguments that venture beyond scientific explanation into theological territory, the most likely explanation is that there are countless universes each with their own fundamental constants and we naturally find ourselves in one compatible with our existence--a selection effect known as the anthropic principle. Eternal inflation offers evidence for this, especially with regard to the cosmological constant. ^[aside]

The "Self-Sampling Assumption" Assumption

Now that we've laid all the necessary groundwork, here is the thesis of this essay: eternal inflation radically changes the set of all observers to which the self-sampling assumption applies. The Fermi paradox and the simulation hypothesis assume we are a random sample from all observers in our bubble universe across all time, when we are actually a random sample from all observers in all bubble universes. ^[aside]

The number of new bubble universes is increasing exponentially over time. Therefore, at any given moment within the broader universe, the vast overwhelming majority of bubble universes have just formed. Likewise, the vast overwhelming majority of intelligent observers will find themselves in the first civilization of their universe. Because by the time a second civilization forms, there will be unfathomably more first civilizations in other universes.

This logic undermines the simulation hypothesis. If the number of real, base-reality bubble universes is growing exponentially and is functionally infinite, then the number of real observers will always vastly outnumber any finite number of simulated observers a civilization might create. Bostrom's trilemma should be a quadrilemma with a fourth possibility: we live in an eternally inflating multiverse.

Let's assume it takes at least T time for a bubble universe to bear intelligent life. Then you'd expect nearly all intelligent observers to exist around T time after the creation of their bubble universe. This would explain why we find ourselves relatively early in the stelliferous era of our universe. It takes time for the universe to cool, for gamma ray bursts or other sterilizing events to die down, for stars to go supernova at just the right size to form neutron stars, for these neutron stars to collide and spew heavy elements, for these heavy elements to coalesce into planets, for these planets to cool, for abiogenesis to occur, and for life to evolve into an intelligent entity capable of asking these questions. Perhaps T = 13.8 billion years.

Remember in the Fermi paradox section when I said that abiogenesis occured essentially the moment that the earth cooled enough for it to happen? I used this as evidence against abiogenesis being rare. But actually, if inflation is true, it's inevitable that this would be the case for nearly all observers regardless of how rare abiogenesis is. Abiogenesis could be so rare that it only occurs in one out of a quadrillion habitable bubble universes and it would still be overwhelmingly likely for an observer to exist at time T. Earth might host the only life that our universe will ever produce, raising the importance of us being a good steward of the planet and preserving the light of consciousness in our universe.

In the case of the Fermi paradox, gazing out at a barren lifeless galaxy is statistical destiny and not a paradox. It is simply the mathematical fact that life capable of asking these questions will almost always arise in a universe too young for any other civilization to have preceded it. The silence of our cosmos is implied by the deafening explosions of infinite Big Bangs.

We are likely the pioneer consciousness of a newborn cosmos. What will we create in this vast and silent dawn? What legacy will the founders of this universe leave behind?